summaryrefslogtreecommitdiffstats
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/Kconfig26
-rw-r--r--fs/btrfs/Makefile2
-rw-r--r--fs/btrfs/backref.c92
-rw-r--r--fs/btrfs/backref.h5
-rw-r--r--fs/btrfs/btrfs_inode.h22
-rw-r--r--fs/btrfs/check-integrity.c3
-rw-r--r--fs/btrfs/compression.c18
-rw-r--r--fs/btrfs/compression.h2
-rw-r--r--fs/btrfs/ctree.c436
-rw-r--r--fs/btrfs/ctree.h295
-rw-r--r--fs/btrfs/delayed-inode.c364
-rw-r--r--fs/btrfs/delayed-inode.h3
-rw-r--r--fs/btrfs/delayed-ref.c112
-rw-r--r--fs/btrfs/delayed-ref.h52
-rw-r--r--fs/btrfs/dev-replace.c6
-rw-r--r--fs/btrfs/dir-item.c11
-rw-r--r--fs/btrfs/disk-io.c656
-rw-r--r--fs/btrfs/disk-io.h12
-rw-r--r--fs/btrfs/export.c4
-rw-r--r--fs/btrfs/extent-tree.c1184
-rw-r--r--fs/btrfs/extent_io.c482
-rw-r--r--fs/btrfs/extent_io.h54
-rw-r--r--fs/btrfs/extent_map.c24
-rw-r--r--fs/btrfs/extent_map.h3
-rw-r--r--fs/btrfs/file-item.c175
-rw-r--r--fs/btrfs/file.c116
-rw-r--r--fs/btrfs/free-space-cache.c658
-rw-r--r--fs/btrfs/free-space-cache.h5
-rw-r--r--fs/btrfs/inode-item.c17
-rw-r--r--fs/btrfs/inode.c1279
-rw-r--r--fs/btrfs/ioctl.c389
-rw-r--r--fs/btrfs/ioctl.h502
-rw-r--r--fs/btrfs/locking.c9
-rw-r--r--fs/btrfs/locking.h1
-rw-r--r--fs/btrfs/ordered-data.c128
-rw-r--r--fs/btrfs/ordered-data.h17
-rw-r--r--fs/btrfs/print-tree.c10
-rw-r--r--fs/btrfs/print-tree.h2
-rw-r--r--fs/btrfs/qgroup.c900
-rw-r--r--fs/btrfs/raid56.c2100
-rw-r--r--fs/btrfs/raid56.h51
-rw-r--r--fs/btrfs/reada.c5
-rw-r--r--fs/btrfs/relocation.c187
-rw-r--r--fs/btrfs/root-tree.c7
-rw-r--r--fs/btrfs/scrub.c143
-rw-r--r--fs/btrfs/send.c95
-rw-r--r--fs/btrfs/send.h2
-rw-r--r--fs/btrfs/super.c183
-rw-r--r--fs/btrfs/sysfs.c1
-rw-r--r--fs/btrfs/transaction.c321
-rw-r--r--fs/btrfs/transaction.h11
-rw-r--r--fs/btrfs/tree-defrag.c19
-rw-r--r--fs/btrfs/tree-log.c571
-rw-r--r--fs/btrfs/tree-log.h3
-rw-r--r--fs/btrfs/ulist.c60
-rw-r--r--fs/btrfs/ulist.h6
-rw-r--r--fs/btrfs/volumes.c736
-rw-r--r--fs/btrfs/volumes.h18
-rw-r--r--fs/btrfs/xattr.c4
59 files changed, 9042 insertions, 3557 deletions
diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index d33f01c08b6..2b3b8329697 100644
--- a/fs/btrfs/Kconfig
+++ b/fs/btrfs/Kconfig
@@ -1,11 +1,13 @@
config BTRFS_FS
- tristate "Btrfs filesystem (EXPERIMENTAL) Unstable disk format"
- depends on EXPERIMENTAL
+ tristate "Btrfs filesystem support"
select LIBCRC32C
select ZLIB_INFLATE
select ZLIB_DEFLATE
select LZO_COMPRESS
select LZO_DECOMPRESS
+ select RAID6_PQ
+ select XOR_BLOCKS
+
help
Btrfs is a new filesystem with extents, writable snapshotting,
support for multiple devices and many more features.
@@ -50,3 +52,23 @@ config BTRFS_FS_CHECK_INTEGRITY
In most cases, unless you are a btrfs developer who needs
to verify the integrity of (super)-block write requests
during the run of a regression test, say N
+
+config BTRFS_FS_RUN_SANITY_TESTS
+ bool "Btrfs will run sanity tests upon loading"
+ depends on BTRFS_FS
+ help
+ This will run some basic sanity tests on the free space cache
+ code to make sure it is acting as it should. These are mostly
+ regression tests and are only really interesting to btrfs devlopers.
+
+ If unsure, say N.
+
+config BTRFS_DEBUG
+ bool "Btrfs debugging support"
+ depends on BTRFS_FS
+ help
+ Enable run-time debugging support for the btrfs filesystem. This may
+ enable additional and expensive checks with negative impact on
+ performance, or export extra information via sysfs.
+
+ If unsure, say N.
diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
index 7df3e0f0ee5..3932224f99e 100644
--- a/fs/btrfs/Makefile
+++ b/fs/btrfs/Makefile
@@ -8,7 +8,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
export.o tree-log.o free-space-cache.o zlib.o lzo.o \
compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \
- reada.o backref.o ulist.o qgroup.o send.o dev-replace.o
+ reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o
btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o
btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 04edf69be87..b4fb4155811 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -352,11 +352,10 @@ static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info,
err = __resolve_indirect_ref(fs_info, search_commit_root,
time_seq, ref, parents,
extent_item_pos);
- if (err) {
- if (ret == 0)
- ret = err;
+ if (err == -ENOMEM)
+ goto out;
+ if (err)
continue;
- }
/* we put the first parent into the ref at hand */
ULIST_ITER_INIT(&uiter);
@@ -370,7 +369,7 @@ static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info,
new_ref = kmalloc(sizeof(*new_ref), GFP_NOFS);
if (!new_ref) {
ret = -ENOMEM;
- break;
+ goto out;
}
memcpy(new_ref, ref, sizeof(*ref));
new_ref->parent = node->val;
@@ -380,7 +379,7 @@ static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info,
}
ulist_reinit(parents);
}
-
+out:
ulist_free(parents);
return ret;
}
@@ -424,7 +423,10 @@ static int __add_missing_keys(struct btrfs_fs_info *fs_info,
BUG_ON(!ref->wanted_disk_byte);
eb = read_tree_block(fs_info->tree_root, ref->wanted_disk_byte,
fs_info->tree_root->leafsize, 0);
- BUG_ON(!eb);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
+ return -EIO;
+ }
btrfs_tree_read_lock(eb);
if (btrfs_header_level(eb) == 0)
btrfs_item_key_to_cpu(eb, &ref->key_for_search, 0);
@@ -446,7 +448,7 @@ static int __add_missing_keys(struct btrfs_fs_info *fs_info,
* having a parent).
* mode = 2: merge identical parents
*/
-static int __merge_refs(struct list_head *head, int mode)
+static void __merge_refs(struct list_head *head, int mode)
{
struct list_head *pos1;
@@ -492,7 +494,6 @@ static int __merge_refs(struct list_head *head, int mode)
}
}
- return 0;
}
/*
@@ -585,7 +586,8 @@ static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq,
default:
WARN_ON(1);
}
- BUG_ON(ret);
+ if (ret)
+ return ret;
}
return 0;
@@ -683,7 +685,8 @@ static int __add_inline_refs(struct btrfs_fs_info *fs_info,
default:
WARN_ON(1);
}
- BUG_ON(ret);
+ if (ret)
+ return ret;
ptr += btrfs_extent_inline_ref_size(type);
}
@@ -765,7 +768,9 @@ static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
default:
WARN_ON(1);
}
- BUG_ON(ret);
+ if (ret)
+ return ret;
+
}
return ret;
@@ -883,18 +888,14 @@ again:
if (ret)
goto out;
- ret = __merge_refs(&prefs, 1);
- if (ret)
- goto out;
+ __merge_refs(&prefs, 1);
ret = __resolve_indirect_refs(fs_info, search_commit_root, time_seq,
&prefs, extent_item_pos);
if (ret)
goto out;
- ret = __merge_refs(&prefs, 2);
- if (ret)
- goto out;
+ __merge_refs(&prefs, 2);
while (!list_empty(&prefs)) {
ref = list_first_entry(&prefs, struct __prelim_ref, list);
@@ -903,7 +904,8 @@ again:
if (ref->count && ref->root_id && ref->parent == 0) {
/* no parent == root of tree */
ret = ulist_add(roots, ref->root_id, 0, GFP_NOFS);
- BUG_ON(ret < 0);
+ if (ret < 0)
+ goto out;
}
if (ref->count && ref->parent) {
struct extent_inode_elem *eie = NULL;
@@ -914,7 +916,10 @@ again:
info_level);
eb = read_tree_block(fs_info->extent_root,
ref->parent, bsz, 0);
- BUG_ON(!eb);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
+ return -EIO;
+ }
ret = find_extent_in_eb(eb, bytenr,
*extent_item_pos, &eie);
ref->inode_list = eie;
@@ -923,6 +928,8 @@ again:
ret = ulist_add_merge(refs, ref->parent,
(uintptr_t)ref->inode_list,
(u64 *)&eie, GFP_NOFS);
+ if (ret < 0)
+ goto out;
if (!ret && extent_item_pos) {
/*
* we've recorded that parent, so we must extend
@@ -933,7 +940,6 @@ again:
eie = eie->next;
eie->next = ref->inode_list;
}
- BUG_ON(ret < 0);
}
kfree(ref);
}
@@ -1183,6 +1189,20 @@ int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
return ret;
}
+/*
+ * this iterates to turn a name (from iref/extref) into a full filesystem path.
+ * Elements of the path are separated by '/' and the path is guaranteed to be
+ * 0-terminated. the path is only given within the current file system.
+ * Therefore, it never starts with a '/'. the caller is responsible to provide
+ * "size" bytes in "dest". the dest buffer will be filled backwards. finally,
+ * the start point of the resulting string is returned. this pointer is within
+ * dest, normally.
+ * in case the path buffer would overflow, the pointer is decremented further
+ * as if output was written to the buffer, though no more output is actually
+ * generated. that way, the caller can determine how much space would be
+ * required for the path to fit into the buffer. in that case, the returned
+ * value will be smaller than dest. callers must check this!
+ */
char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
u32 name_len, unsigned long name_off,
struct extent_buffer *eb_in, u64 parent,
@@ -1252,32 +1272,6 @@ char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
}
/*
- * this iterates to turn a btrfs_inode_ref into a full filesystem path. elements
- * of the path are separated by '/' and the path is guaranteed to be
- * 0-terminated. the path is only given within the current file system.
- * Therefore, it never starts with a '/'. the caller is responsible to provide
- * "size" bytes in "dest". the dest buffer will be filled backwards. finally,
- * the start point of the resulting string is returned. this pointer is within
- * dest, normally.
- * in case the path buffer would overflow, the pointer is decremented further
- * as if output was written to the buffer, though no more output is actually
- * generated. that way, the caller can determine how much space would be
- * required for the path to fit into the buffer. in that case, the returned
- * value will be smaller than dest. callers must check this!
- */
-char *btrfs_iref_to_path(struct btrfs_root *fs_root,
- struct btrfs_path *path,
- struct btrfs_inode_ref *iref,
- struct extent_buffer *eb_in, u64 parent,
- char *dest, u32 size)
-{
- return btrfs_ref_to_path(fs_root, path,
- btrfs_inode_ref_name_len(eb_in, iref),
- (unsigned long)(iref + 1),
- eb_in, parent, dest, size);
-}
-
-/*
* this makes the path point to (logical EXTENT_ITEM *)
* returns BTRFS_EXTENT_FLAG_DATA for data, BTRFS_EXTENT_FLAG_TREE_BLOCK for
* tree blocks and <0 on error.
@@ -1464,8 +1458,6 @@ int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
iterate_extent_inodes_t *iterate, void *ctx)
{
int ret;
- struct list_head data_refs = LIST_HEAD_INIT(data_refs);
- struct list_head shared_refs = LIST_HEAD_INIT(shared_refs);
struct btrfs_trans_handle *trans;
struct ulist *refs = NULL;
struct ulist *roots = NULL;
@@ -1511,11 +1503,9 @@ int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
iterate, ctx);
}
ulist_free(roots);
- roots = NULL;
}
free_leaf_list(refs);
- ulist_free(roots);
out:
if (!search_commit_root) {
btrfs_put_tree_mod_seq(fs_info, &tree_mod_seq_elem);
diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h
index d61feca7945..0f446d7ca2c 100644
--- a/fs/btrfs/backref.h
+++ b/fs/btrfs/backref.h
@@ -19,7 +19,7 @@
#ifndef __BTRFS_BACKREF__
#define __BTRFS_BACKREF__
-#include "ioctl.h"
+#include <linux/btrfs.h>
#include "ulist.h"
#include "extent_io.h"
@@ -59,9 +59,6 @@ int paths_from_inode(u64 inum, struct inode_fs_paths *ipath);
int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 bytenr,
u64 time_seq, struct ulist **roots);
-char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
- struct btrfs_inode_ref *iref, struct extent_buffer *eb,
- u64 parent, char *dest, u32 size);
char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
u32 name_len, unsigned long name_off,
struct extent_buffer *eb_in, u64 parent,
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 2a8c242bc4f..08b286b2a2c 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -40,6 +40,8 @@
#define BTRFS_INODE_HAS_ASYNC_EXTENT 6
#define BTRFS_INODE_NEEDS_FULL_SYNC 7
#define BTRFS_INODE_COPY_EVERYTHING 8
+#define BTRFS_INODE_IN_DELALLOC_LIST 9
+#define BTRFS_INODE_READDIO_NEED_LOCK 10
/* in memory btrfs inode */
struct btrfs_inode {
@@ -91,7 +93,7 @@ struct btrfs_inode {
unsigned long runtime_flags;
- /* Keep track of who's O_SYNC/fsycing currently */
+ /* Keep track of who's O_SYNC/fsyncing currently */
atomic_t sync_writers;
/* full 64 bit generation number, struct vfs_inode doesn't have a big
@@ -216,4 +218,22 @@ static inline int btrfs_inode_in_log(struct inode *inode, u64 generation)
return 0;
}
+/*
+ * Disable DIO read nolock optimization, so new dio readers will be forced
+ * 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)
+{
+ set_bit(BTRFS_INODE_READDIO_NEED_LOCK, &BTRFS_I(inode)->runtime_flags);
+ smp_mb();
+}
+
+static inline void btrfs_inode_resume_unlocked_dio(struct inode *inode)
+{
+ smp_mb__before_clear_bit();
+ clear_bit(BTRFS_INODE_READDIO_NEED_LOCK,
+ &BTRFS_I(inode)->runtime_flags);
+}
+
#endif
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index 11d47bfb62b..18af6f48781 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -813,8 +813,7 @@ static int btrfsic_process_superblock_dev_mirror(
(bh->b_data + (dev_bytenr & 4095));
if (btrfs_super_bytenr(super_tmp) != dev_bytenr ||
- strncmp((char *)(&(super_tmp->magic)), BTRFS_MAGIC,
- sizeof(super_tmp->magic)) ||
+ super_tmp->magic != cpu_to_le64(BTRFS_MAGIC) ||
memcmp(device->uuid, super_tmp->dev_item.uuid, BTRFS_UUID_SIZE) ||
btrfs_super_nodesize(super_tmp) != state->metablock_size ||
btrfs_super_leafsize(super_tmp) != state->metablock_size ||
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 94ab2f80e7e..b189bd1e7a3 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -82,6 +82,10 @@ struct compressed_bio {
u32 sums;
};
+static int btrfs_decompress_biovec(int type, struct page **pages_in,
+ u64 disk_start, struct bio_vec *bvec,
+ int vcnt, size_t srclen);
+
static inline int compressed_bio_size(struct btrfs_root *root,
unsigned long disk_size)
{
@@ -106,7 +110,6 @@ static int check_compressed_csum(struct inode *inode,
u64 disk_start)
{
int ret;
- struct btrfs_root *root = BTRFS_I(inode)->root;
struct page *page;
unsigned long i;
char *kaddr;
@@ -121,7 +124,7 @@ static int check_compressed_csum(struct inode *inode,
csum = ~(u32)0;
kaddr = kmap_atomic(page);
- csum = btrfs_csum_data(root, kaddr, csum, PAGE_CACHE_SIZE);
+ csum = btrfs_csum_data(kaddr, csum, PAGE_CACHE_SIZE);
btrfs_csum_final(csum, (char *)&csum);
kunmap_atomic(kaddr);
@@ -372,7 +375,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
page = compressed_pages[pg_index];
page->mapping = inode->i_mapping;
if (bio->bi_size)
- ret = io_tree->ops->merge_bio_hook(page, 0,
+ ret = io_tree->ops->merge_bio_hook(WRITE, page, 0,
PAGE_CACHE_SIZE,
bio, 0);
else
@@ -655,7 +658,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
page->index = em_start >> PAGE_CACHE_SHIFT;
if (comp_bio->bi_size)
- ret = tree->ops->merge_bio_hook(page, 0,
+ ret = tree->ops->merge_bio_hook(READ, page, 0,
PAGE_CACHE_SIZE,
comp_bio, 0);
else
@@ -739,7 +742,7 @@ static int comp_num_workspace[BTRFS_COMPRESS_TYPES];
static atomic_t comp_alloc_workspace[BTRFS_COMPRESS_TYPES];
static wait_queue_head_t comp_workspace_wait[BTRFS_COMPRESS_TYPES];
-struct btrfs_compress_op *btrfs_compress_op[] = {
+static struct btrfs_compress_op *btrfs_compress_op[] = {
&btrfs_zlib_compress,
&btrfs_lzo_compress,
};
@@ -910,8 +913,9 @@ 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.
*/
-int btrfs_decompress_biovec(int type, struct page **pages_in, u64 disk_start,
- struct bio_vec *bvec, int vcnt, size_t srclen)
+static int btrfs_decompress_biovec(int type, struct page **pages_in,
+ u64 disk_start, struct bio_vec *bvec,
+ int vcnt, size_t srclen)
{
struct list_head *workspace;
int ret;
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 9afb0a62ae8..0c803b4fbf9 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -30,8 +30,6 @@ int btrfs_compress_pages(int type, struct address_space *mapping,
unsigned long *total_in,
unsigned long *total_out,
unsigned long max_out);
-int btrfs_decompress_biovec(int type, struct page **pages_in, u64 disk_start,
- struct bio_vec *bvec, int vcnt, size_t srclen);
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,
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index eea5da7a2b9..de6de8e60b4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -37,16 +37,11 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct extent_buffer *dst_buf,
struct extent_buffer *src_buf);
-static void del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct btrfs_path *path, int level, int slot);
+static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
+ int level, int slot);
static void tree_mod_log_free_eb(struct btrfs_fs_info *fs_info,
struct extent_buffer *eb);
-struct extent_buffer *read_old_tree_block(struct btrfs_root *root, u64 bytenr,
- u32 blocksize, u64 parent_transid,
- u64 time_seq);
-struct extent_buffer *btrfs_find_old_tree_block(struct btrfs_root *root,
- u64 bytenr, u32 blocksize,
- u64 time_seq);
+static int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path);
struct btrfs_path *btrfs_alloc_path(void)
{
@@ -208,7 +203,7 @@ struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
* tree until you end up with a lock on the root. A locked buffer
* is returned, with a reference held.
*/
-struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
+static struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
{
struct extent_buffer *eb;
@@ -361,6 +356,44 @@ static inline void tree_mod_log_write_unlock(struct btrfs_fs_info *fs_info)
}
/*
+ * Increment the upper half of tree_mod_seq, set lower half zero.
+ *
+ * Must be called with fs_info->tree_mod_seq_lock held.
+ */
+static inline u64 btrfs_inc_tree_mod_seq_major(struct btrfs_fs_info *fs_info)
+{
+ u64 seq = atomic64_read(&fs_info->tree_mod_seq);
+ seq &= 0xffffffff00000000ull;
+ seq += 1ull << 32;
+ atomic64_set(&fs_info->tree_mod_seq, seq);
+ return seq;
+}
+
+/*
+ * Increment the lower half of tree_mod_seq.
+ *
+ * Must be called with fs_info->tree_mod_seq_lock held. The way major numbers
+ * are generated should not technically require a spin lock here. (Rationale:
+ * incrementing the minor while incrementing the major seq number is between its
+ * atomic64_read and atomic64_set calls doesn't duplicate sequence numbers, it
+ * just returns a unique sequence number as usual.) We have decided to leave
+ * that requirement in here and rethink it once we notice it really imposes a
+ * problem on some workload.
+ */
+static inline u64 btrfs_inc_tree_mod_seq_minor(struct btrfs_fs_info *fs_info)
+{
+ return atomic64_inc_return(&fs_info->tree_mod_seq);
+}
+
+/*
+ * return the last minor in the previous major tree_mod_seq number
+ */
+u64 btrfs_tree_mod_seq_prev(u64 seq)
+{
+ return (seq & 0xffffffff00000000ull) - 1ull;
+}
+
+/*
* This adds a new blocker to the tree mod log's blocker list if the @elem
* passed does not already have a sequence number set. So when a caller expects
* to record tree modifications, it should ensure to set elem->seq to zero
@@ -376,10 +409,10 @@ u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
tree_mod_log_write_lock(fs_info);
spin_lock(&fs_info->tree_mod_seq_lock);
if (!elem->seq) {
- elem->seq = btrfs_inc_tree_mod_seq(fs_info);
+ elem->seq = btrfs_inc_tree_mod_seq_major(fs_info);
list_add_tail(&elem->list, &fs_info->tree_mod_seq_list);
}
- seq = btrfs_inc_tree_mod_seq(fs_info);
+ seq = btrfs_inc_tree_mod_seq_minor(fs_info);
spin_unlock(&fs_info->tree_mod_seq_lock);
tree_mod_log_write_unlock(fs_info);
@@ -524,7 +557,10 @@ static inline int tree_mod_alloc(struct btrfs_fs_info *fs_info, gfp_t flags,
if (!tm)
return -ENOMEM;
- tm->seq = btrfs_inc_tree_mod_seq(fs_info);
+ spin_lock(&fs_info->tree_mod_seq_lock);
+ tm->seq = btrfs_inc_tree_mod_seq_minor(fs_info);
+ spin_unlock(&fs_info->tree_mod_seq_lock);
+
return tm->seq;
}
@@ -643,7 +679,8 @@ __tree_mod_log_free_eb(struct btrfs_fs_info *fs_info, struct extent_buffer *eb)
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, gfp_t flags,
+ int log_removal)
{
struct tree_mod_elem *tm;
int ret;
@@ -651,6 +688,9 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
if (tree_mod_dont_log(fs_info, NULL))
return 0;
+ if (log_removal)
+ __tree_mod_log_free_eb(fs_info, old_root);
+
ret = tree_mod_alloc(fs_info, flags, &tm);
if (ret < 0)
goto out;
@@ -751,8 +791,8 @@ tree_mod_log_eb_copy(struct btrfs_fs_info *fs_info, struct extent_buffer *dst,
for (i = 0; i < nr_items; i++) {
ret = tree_mod_log_insert_key_locked(fs_info, src,
- i + src_offset,
- MOD_LOG_KEY_REMOVE);
+ i + src_offset,
+ MOD_LOG_KEY_REMOVE);
BUG_ON(ret < 0);
ret = tree_mod_log_insert_key_locked(fs_info, dst,
i + dst_offset,
@@ -798,11 +838,12 @@ tree_mod_log_free_eb(struct btrfs_fs_info *fs_info, struct extent_buffer *eb)
static noinline void
tree_mod_log_set_root_pointer(struct btrfs_root *root,
- struct extent_buffer *new_root_node)
+ struct extent_buffer *new_root_node,
+ int log_removal)
{
int ret;
ret = tree_mod_log_insert_root(root->fs_info, root->node,
- new_root_node, GFP_NOFS);
+ new_root_node, GFP_NOFS, log_removal);
BUG_ON(ret < 0);
}
@@ -863,7 +904,8 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
if (btrfs_block_can_be_shared(root, buf)) {
ret = btrfs_lookup_extent_info(trans, root, buf->start,
- buf->len, &refs, &flags);
+ btrfs_header_level(buf), 1,
+ &refs, &flags);
if (ret)
return ret;
if (refs == 0) {
@@ -927,7 +969,6 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
ret = btrfs_dec_ref(trans, root, buf, 1, 1);
BUG_ON(ret); /* -ENOMEM */
}
- tree_mod_log_free_eb(root->fs_info, buf);
clean_tree_block(trans, root, buf);
*last_ref = 1;
}
@@ -1025,7 +1066,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
parent_start = 0;
extent_buffer_get(cow);
- tree_mod_log_set_root_pointer(root, cow);
+ tree_mod_log_set_root_pointer(root, cow, 1);
rcu_assign_pointer(root->node, cow);
btrfs_free_tree_block(trans, root, buf, parent_start,
@@ -1046,6 +1087,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
btrfs_set_node_ptr_generation(parent, parent_slot,
trans->transid);
btrfs_mark_buffer_dirty(parent);
+ tree_mod_log_free_eb(root->fs_info, buf);
btrfs_free_tree_block(trans, root, buf, parent_start,
last_ref);
}
@@ -1063,11 +1105,11 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
*/
static struct tree_mod_elem *
__tree_mod_log_oldest_root(struct btrfs_fs_info *fs_info,
- struct btrfs_root *root, u64 time_seq)
+ struct extent_buffer *eb_root, u64 time_seq)
{
struct tree_mod_elem *tm;
struct tree_mod_elem *found = NULL;
- u64 root_logical = root->node->start;
+ u64 root_logical = eb_root->start;
int looped = 0;
if (!time_seq)
@@ -1101,7 +1143,6 @@ __tree_mod_log_oldest_root(struct btrfs_fs_info *fs_info,
found = tm;
root_logical = tm->old_root.logical;
- BUG_ON(root_logical == root->node->start);
looped = 1;
}
@@ -1138,6 +1179,7 @@ __tree_mod_log_rewind(struct extent_buffer *eb, u64 time_seq,
switch (tm->op) {
case MOD_LOG_KEY_REMOVE_WHILE_FREEING:
BUG_ON(tm->slot < n);
+ /* Fallthrough */
case MOD_LOG_KEY_REMOVE_WHILE_MOVING:
case MOD_LOG_KEY_REMOVE:
btrfs_set_node_key(eb, &tm->key, tm->slot);
@@ -1185,6 +1227,13 @@ __tree_mod_log_rewind(struct extent_buffer *eb, u64 time_seq,
btrfs_set_header_nritems(eb, n);
}
+/*
+ * Called with eb read locked. If the buffer cannot be rewinded, the same buffer
+ * is returned. If rewind operations happen, a fresh buffer is returned. The
+ * returned buffer is always read-locked. If the returned buffer is not the
+ * input buffer, the lock on the input buffer is released and the input buffer
+ * is freed (its refcount is decremented).
+ */
static struct extent_buffer *
tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
u64 time_seq)
@@ -1218,11 +1267,14 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
}
extent_buffer_get(eb_rewin);
+ btrfs_tree_read_unlock(eb);
free_extent_buffer(eb);
+ extent_buffer_get(eb_rewin);
+ btrfs_tree_read_lock(eb_rewin);
__tree_mod_log_rewind(eb_rewin, time_seq, tm);
WARN_ON(btrfs_header_nritems(eb_rewin) >
- BTRFS_NODEPTRS_PER_BLOCK(fs_info->fs_root));
+ BTRFS_NODEPTRS_PER_BLOCK(fs_info->tree_root));
return eb_rewin;
}
@@ -1238,33 +1290,35 @@ static inline struct extent_buffer *
get_old_root(struct btrfs_root *root, u64 time_seq)
{
struct tree_mod_elem *tm;
- struct extent_buffer *eb;
+ struct extent_buffer *eb = NULL;
+ struct extent_buffer *eb_root;
struct extent_buffer *old;
struct tree_mod_root *old_root = NULL;
u64 old_generation = 0;
u64 logical;
u32 blocksize;
- eb = btrfs_read_lock_root_node(root);
- tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq);
+ eb_root = btrfs_read_lock_root_node(root);
+ tm = __tree_mod_log_oldest_root(root->fs_info, eb_root, time_seq);
if (!tm)
- return root->node;
+ return eb_root;
if (tm->op == MOD_LOG_ROOT_REPLACE) {
old_root = &tm->old_root;
old_generation = tm->generation;
logical = old_root->logical;
} else {
- logical = root->node->start;
+ logical = eb_root->start;
}
tm = tree_mod_log_search(root->fs_info, logical, time_seq);
if (old_root && tm && tm->op != MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
- btrfs_tree_read_unlock(root->node);
- free_extent_buffer(root->node);
+ btrfs_tree_read_unlock(eb_root);
+ free_extent_buffer(eb_root);
blocksize = btrfs_level_size(root, old_root->level);
old = read_tree_block(root, logical, blocksize, 0);
- if (!old) {
+ if (!old || !extent_buffer_uptodate(old)) {
+ free_extent_buffer(old);
pr_warn("btrfs: failed to read tree block %llu from get_old_root\n",
logical);
WARN_ON(1);
@@ -1273,13 +1327,13 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
free_extent_buffer(old);
}
} else if (old_root) {
- btrfs_tree_read_unlock(root->node);
- free_extent_buffer(root->node);
+ btrfs_tree_read_unlock(eb_root);
+ free_extent_buffer(eb_root);
eb = alloc_dummy_extent_buffer(logical, root->nodesize);
} else {
- eb = btrfs_clone_extent_buffer(root->node);
- btrfs_tree_read_unlock(root->node);
- free_extent_buffer(root->node);
+ eb = btrfs_clone_extent_buffer(eb_root);
+ btrfs_tree_read_unlock(eb_root);
+ free_extent_buffer(eb_root);
}
if (!eb)
@@ -1289,7 +1343,7 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
if (old_root) {
btrfs_set_header_bytenr(eb, eb->start);
btrfs_set_header_backref_rev(eb, BTRFS_MIXED_BACKREF_REV);
- btrfs_set_header_owner(eb, root->root_key.objectid);
+ btrfs_set_header_owner(eb, btrfs_header_owner(eb_root));
btrfs_set_header_level(eb, old_root->level);
btrfs_set_header_generation(eb, old_generation);
}
@@ -1306,15 +1360,15 @@ int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq)
{
struct tree_mod_elem *tm;
int level;
+ struct extent_buffer *eb_root = btrfs_root_node(root);
- tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq);
+ tm = __tree_mod_log_oldest_root(root->fs_info, eb_root, time_seq);
if (tm && tm->op == MOD_LOG_ROOT_REPLACE) {
level = tm->old_root.level;
} else {
- rcu_read_lock();
- level = btrfs_header_level(root->node);
- rcu_read_unlock();
+ level = btrfs_header_level(eb_root);
}
+ free_extent_buffer(eb_root);
return level;
}
@@ -1441,7 +1495,7 @@ int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2)
*/
int btrfs_realloc_node(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct extent_buffer *parent,
- int start_slot, int cache_only, u64 *last_ret,
+ int start_slot, u64 *last_ret,
struct btrfs_key *progress)
{
struct extent_buffer *cur;
@@ -1461,8 +1515,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
struct btrfs_disk_key disk_key;
parent_level = btrfs_header_level(parent);
- if (cache_only && parent_level != 1)
- return 0;
WARN_ON(trans->transaction != root->fs_info->running_transaction);
WARN_ON(trans->transid != root->fs_info->generation);
@@ -1508,15 +1560,13 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
else
uptodate = 0;
if (!cur || !uptodate) {
- if (cache_only) {
- free_extent_buffer(cur);
- continue;
- }
if (!cur) {
cur = read_tree_block(root, blocknr,
blocksize, gen);
- if (!cur)
+ if (!cur || !extent_buffer_uptodate(cur)) {
+ free_extent_buffer(cur);
return -EIO;
+ }
} else if (!uptodate) {
err = btrfs_read_buffer(cur, gen);
if (err) {
@@ -1681,6 +1731,8 @@ static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root,
struct extent_buffer *parent, int slot)
{
int level = btrfs_header_level(parent);
+ struct extent_buffer *eb;
+
if (slot < 0)
return NULL;
if (slot >= btrfs_header_nritems(parent))
@@ -1688,9 +1740,15 @@ static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root,
BUG_ON(level == 0);
- return read_tree_block(root, btrfs_node_blockptr(parent, slot),
- btrfs_level_size(root, level - 1),
- btrfs_node_ptr_generation(parent, slot));
+ eb = read_tree_block(root, btrfs_node_blockptr(parent, slot),
+ btrfs_level_size(root, level - 1),
+ btrfs_node_ptr_generation(parent, slot));
+ if (eb && !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
+ eb = NULL;
+ }
+
+ return eb;
}
/*
@@ -1755,8 +1813,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
goto enospc;
}
- tree_mod_log_free_eb(root->fs_info, root->node);
- tree_mod_log_set_root_pointer(root, child);
+ tree_mod_log_set_root_pointer(root, child, 1);
rcu_assign_pointer(root->node, child);
add_root_to_dirty_list(root);
@@ -1820,7 +1877,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
if (btrfs_header_nritems(right) == 0) {
clean_tree_block(trans, root, right);
btrfs_tree_unlock(right);
- del_ptr(trans, root, path, level + 1, pslot + 1);
+ del_ptr(root, path, level + 1, pslot + 1);
root_sub_used(root, right->len);
btrfs_free_tree_block(trans, root, right, 0, 1);
free_extent_buffer_stale(right);
@@ -1864,7 +1921,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
if (btrfs_header_nritems(mid) == 0) {
clean_tree_block(trans, root, mid);
btrfs_tree_unlock(mid);
- del_ptr(trans, root, path, level + 1, pslot);
+ del_ptr(root, path, level + 1, pslot);
root_sub_used(root, mid->len);
btrfs_free_tree_block(trans, root, mid, 0, 1);
free_extent_buffer_stale(mid);
@@ -2212,9 +2269,6 @@ static noinline void unlock_up(struct btrfs_path *path, int level,
int no_skips = 0;
struct extent_buffer *t;
- if (path->really_keep_locks)
- return;
-
for (i = level; i < BTRFS_MAX_LEVEL; i++) {
if (!path->nodes[i])
break;
@@ -2262,7 +2316,7 @@ noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
{
int i;
- if (path->keep_locks || path->really_keep_locks)
+ if (path->keep_locks)
return;
for (i = level; i < BTRFS_MAX_LEVEL; i++) {
@@ -2495,7 +2549,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
if (!cow)
write_lock_level = -1;
- if (cow && (p->really_keep_locks || p->keep_locks || p->lowest_level))
+ if (cow && (p->keep_locks || p->lowest_level))
write_lock_level = BTRFS_MAX_LEVEL;
min_write_lock_level = write_lock_level;
@@ -2797,15 +2851,9 @@ again:
btrfs_clear_path_blocking(p, b,
BTRFS_READ_LOCK);
}
+ b = tree_mod_log_rewind(root->fs_info, b, time_seq);
p->locks[level] = BTRFS_READ_LOCK;
p->nodes[level] = b;
- b = tree_mod_log_rewind(root->fs_info, b, time_seq);
- if (b != p->nodes[level]) {
- btrfs_tree_unlock_rw(p->nodes[level],
- p->locks[level]);
- p->locks[level] = 0;
- p->nodes[level] = b;
- }
} else {
p->slots[level] = slot;
unlock_up(p, level, lowest_unlock, 0, NULL);
@@ -2904,8 +2952,7 @@ again:
* higher levels
*
*/
-static void fixup_low_keys(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+static void fixup_low_keys(struct btrfs_root *root, struct btrfs_path *path,
struct btrfs_disk_key *key, int level)
{
int i;
@@ -2930,8 +2977,7 @@ static void fixup_low_keys(struct btrfs_trans_handle *trans,
* This function isn't completely safe. It's the caller's responsibility
* that the new key won't break the order
*/
-void btrfs_set_item_key_safe(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void btrfs_set_item_key_safe(struct btrfs_root *root, struct btrfs_path *path,
struct btrfs_key *new_key)
{
struct btrfs_disk_key disk_key;
@@ -2953,7 +2999,7 @@ void btrfs_set_item_key_safe(struct btrfs_trans_handle *trans,
btrfs_set_item_key(eb, &disk_key, slot);
btrfs_mark_buffer_dirty(eb);
if (slot == 0)
- fixup_low_keys(trans, root, path, &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
}
/*
@@ -3095,7 +3141,7 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
*/
static noinline int insert_new_root(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct btrfs_path *path, int level)
+ struct btrfs_path *path, int level, int log_removal)
{
u64 lower_gen;
struct extent_buffer *lower;
@@ -3146,7 +3192,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
btrfs_mark_buffer_dirty(c);
old = root->node;
- tree_mod_log_set_root_pointer(root, c);
+ tree_mod_log_set_root_pointer(root, c, log_removal);
rcu_assign_pointer(root->node, c);
/* the super has an extra ref to root->node */
@@ -3227,8 +3273,17 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
c = path->nodes[level];
WARN_ON(btrfs_header_generation(c) != trans->transid);
if (c == root->node) {
- /* trying to split the root, lets make a new one */
- ret = insert_new_root(trans, root, path, level + 1);
+ /*
+ * trying to split the root, lets make a new one
+ *
+ * tree mod log: We pass 0 as log_removal parameter to
+ * insert_new_root, because that root buffer will be kept as a
+ * normal node. We are going to log removal of half of the
+ * elements below with tree_mod_log_eb_copy. We're holding a
+ * tree lock on the buffer, which is why we cannot race with
+ * other tree_mod_log users.
+ */
+ ret = insert_new_root(trans, root, path, level + 1, 0);
if (ret)
return ret;
} else {
@@ -3682,7 +3737,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
clean_tree_block(trans, root, right);
btrfs_item_key(right, &disk_key, 0);
- fixup_low_keys(trans, root, path, &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
/* then fixup the leaf pointer in the path */
if (path->slots[0] < push_items) {
@@ -3948,7 +4003,7 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
}
if (!path->nodes[1]) {
- ret = insert_new_root(trans, root, path, 1);
+ ret = insert_new_root(trans, root, path, 1, 1);
if (ret)
return ret;
}
@@ -4042,8 +4097,7 @@ again:
path->nodes[0] = right;
path->slots[0] = 0;
if (path->slots[1] == 0)
- fixup_low_keys(trans, root, path,
- &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
}
btrfs_mark_buffer_dirty(right);
return ret;
@@ -4259,7 +4313,7 @@ int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
return ret;
path->slots[0]++;
- setup_items_for_insert(trans, root, path, new_key, &item_size,
+ setup_items_for_insert(root, path, new_key, &item_size,
item_size, item_size +
sizeof(struct btrfs_item), 1);
leaf = path->nodes[0];
@@ -4276,9 +4330,7 @@ int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
* off the end of the item or if we shift the item to chop bytes off
* the front.
*/
-void btrfs_truncate_item(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
+void btrfs_truncate_item(struct btrfs_root *root, struct btrfs_path *path,
u32 new_size, int from_end)
{
int slot;
@@ -4362,7 +4414,7 @@ void btrfs_truncate_item(struct btrfs_trans_handle *trans,
btrfs_set_disk_key_offset(&disk_key, offset + size_diff);
btrfs_set_item_key(leaf, &disk_key, slot);
if (slot == 0)
- fixup_low_keys(trans, root, path, &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
}
item = btrfs_item_nr(leaf, slot);
@@ -4378,8 +4430,7 @@ void btrfs_truncate_item(struct btrfs_trans_handle *trans,
/*
* make the item pointed to by the path bigger, data_size is the new size.
*/
-void btrfs_extend_item(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void btrfs_extend_item(struct btrfs_root *root, struct btrfs_path *path,
u32 data_size)
{
int slot;
@@ -4449,8 +4500,7 @@ void btrfs_extend_item(struct btrfs_trans_handle *trans,
* to save stack depth by doing the bulk of the work in a function
* that doesn't call btrfs_search_slot
*/
-void setup_items_for_insert(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
struct btrfs_key *cpu_key, u32 *data_size,
u32 total_data, u32 total_size, int nr)
{
@@ -4526,7 +4576,7 @@ void setup_items_for_insert(struct btrfs_trans_handle *trans,
if (slot == 0) {
btrfs_cpu_key_to_disk(&disk_key, cpu_key);
- fixup_low_keys(trans, root, path, &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
}
btrfs_unlock_up_safe(path, 1);
btrfs_mark_buffer_dirty(leaf);
@@ -4566,7 +4616,7 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
slot = path->slots[0];
BUG_ON(slot < 0);
- setup_items_for_insert(trans, root, path, cpu_key, data_size,
+ setup_items_for_insert(root, path, cpu_key, data_size,
total_data, total_size, nr);
return 0;
}
@@ -4604,8 +4654,8 @@ int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
* the tree should have been previously balanced so the deletion does not
* empty a node.
*/
-static void del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct btrfs_path *path, int level, int slot)
+static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
+ int level, int slot)
{
struct extent_buffer *parent = path->nodes[level];
u32 nritems;
@@ -4637,7 +4687,7 @@ static void del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
struct btrfs_disk_key disk_key;
btrfs_node_key(parent, &disk_key, 0);
- fixup_low_keys(trans, root, path, &disk_key, level + 1);
+ fixup_low_keys(root, path, &disk_key, level + 1);
}
btrfs_mark_buffer_dirty(parent);
}
@@ -4658,7 +4708,7 @@ static noinline void btrfs_del_leaf(struct btrfs_trans_handle *trans,
struct extent_buffer *leaf)
{
WARN_ON(btrfs_header_generation(leaf) != trans->transid);
- del_ptr(trans, root, path, 1, path->slots[1]);
+ del_ptr(root, path, 1, path->slots[1]);
/*
* btrfs_free_extent is expensive, we want to make sure we
@@ -4739,7 +4789,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
struct btrfs_disk_key disk_key;
btrfs_item_key(leaf, &disk_key, 0);
- fixup_low_keys(trans, root, path, &disk_key, 1);
+ fixup_low_keys(root, path, &disk_key, 1);
}
/* delete the leaf if it is mostly empty */
@@ -4825,8 +4875,8 @@ int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
/*
* A helper function to walk down the tree starting at min_key, and looking
- * for nodes or leaves that are either in cache or have a minimum
- * transaction id. This is used by the btree defrag code, and tree logging
+ * for nodes or leaves that are have a minimum transaction id.
+ * This is used by the btree defrag code, and tree logging
*
* This does not cow, but it does stuff the starting key it finds back
* into min_key, so you can call btrfs_search_slot with cow=1 on the
@@ -4847,7 +4897,7 @@ int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
*/
int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
struct btrfs_key *max_key,
- struct btrfs_path *path, int cache_only,
+ struct btrfs_path *path,
u64 min_trans)
{
struct extent_buffer *cur;
@@ -4887,15 +4937,12 @@ again:
if (sret && slot > 0)
slot--;
/*
- * check this node pointer against the cache_only and
- * min_trans parameters. If it isn't in cache or is too
- * old, skip to the next one.
+ * check this node pointer against the min_trans parameters.
+ * If it is too old, old, skip to the next one.
*/
while (slot < nritems) {
u64 blockptr;
u64 gen;
- struct extent_buffer *tmp;
- struct btrfs_disk_key disk_key;
blockptr = btrfs_node_blockptr(cur, slot);
gen = btrfs_node_ptr_generation(cur, slot);
@@ -4903,27 +4950,7 @@ again:
slot++;
continue;
}
- if (!cache_only)
- break;
-
- if (max_key) {
- btrfs_node_key(cur, &disk_key, slot);
- if (comp_keys(&disk_key, max_key) >= 0) {
- ret = 1;
- goto out;
- }
- }
-
- tmp = btrfs_find_tree_block(root, blockptr,
- btrfs_level_size(root, level - 1));
-
- if (tmp && btrfs_buffer_uptodate(tmp, gen, 1) > 0) {
- free_extent_buffer(tmp);
- break;
- }
- if (tmp)
- free_extent_buffer(tmp);
- slot++;
+ break;
}
find_next_key:
/*
@@ -4934,7 +4961,7 @@ find_next_key:
path->slots[level] = slot;
btrfs_set_path_blocking(path);
sret = btrfs_find_next_key(root, path, min_key, level,
- cache_only, min_trans);
+ min_trans);
if (sret == 0) {
btrfs_release_path(path);
goto again;
@@ -5399,8 +5426,7 @@ out:
/*
* this is similar to btrfs_next_leaf, but does not try to preserve
* and fixup the path. It looks for and returns the next key in the
- * tree based on the current path and the cache_only and min_trans
- * parameters.
+ * tree based on the current path and the min_trans parameters.
*
* 0 is returned if another key is found, < 0 if there are any errors
* and 1 is returned if there are no higher keys in the tree
@@ -5409,8 +5435,7 @@ out:
* calling this function.
*/
int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
- struct btrfs_key *key, int level,
- int cache_only, u64 min_trans)
+ struct btrfs_key *key, int level, u64 min_trans)
{
int slot;
struct extent_buffer *c;
@@ -5461,22 +5486,8 @@ next:
if (level == 0)
btrfs_item_key_to_cpu(c, key, slot);
else {
- u64 blockptr = btrfs_node_blockptr(c, slot);
u64 gen = btrfs_node_ptr_generation(c, slot);
- if (cache_only) {
- struct extent_buffer *cur;
- cur = btrfs_find_tree_block(root, blockptr,
- btrfs_level_size(root, level - 1));
- if (!cur ||
- btrfs_buffer_uptodate(cur, gen, 1) <= 0) {
- slot++;
- if (cur)
- free_extent_buffer(cur);
- goto next;
- }
- free_extent_buffer(cur);
- }
if (gen < min_trans) {
slot++;
goto next;
@@ -5498,139 +5509,6 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
return btrfs_next_old_leaf(root, path, 0);
}
-/* Release the path up to but not including the given level */
-static void btrfs_release_level(struct btrfs_path *path, int level)
-{
- int i;
-
- for (i = 0; i < level; i++) {
- path->slots[i] = 0;
- if (!path->nodes[i])
- continue;
- if (path->locks[i]) {
- btrfs_tree_unlock_rw(path->nodes[i], path->locks[i]);
- path->locks[i] = 0;
- }
- free_extent_buffer(path->nodes[i]);
- path->nodes[i] = NULL;
- }
-}
-
-/*
- * This function assumes 2 things
- *
- * 1) You are using path->keep_locks
- * 2) You are not inserting items.
- *
- * If either of these are not true do not use this function. If you need a next
- * leaf with either of these not being true then this function can be easily
- * adapted to do that, but at the moment these are the limitations.
- */
-int btrfs_next_leaf_write(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
- int del)
-{
- struct extent_buffer *b;
- struct btrfs_key key;
- u32 nritems;
- int level = 1;
- int slot;
- int ret = 1;
- int write_lock_level = BTRFS_MAX_LEVEL;
- int ins_len = del ? -1 : 0;
-
- WARN_ON(!(path->keep_locks || path->really_keep_locks));
-
- nritems = btrfs_header_nritems(path->nodes[0]);
- btrfs_item_key_to_cpu(path->nodes[0], &key, nritems - 1);
-
- while (path->nodes[level]) {
- nritems = btrfs_header_nritems(path->nodes[level]);
- if (!(path->locks[level] & BTRFS_WRITE_LOCK)) {
-search:
- btrfs_release_path(path);
- ret = btrfs_search_slot(trans, root, &key, path,
- ins_len, 1);
- if (ret < 0)
- goto out;
- level = 1;
- continue;
- }
-
- if (path->slots[level] >= nritems - 1) {
- level++;
- continue;
- }
-
- btrfs_release_level(path, level);
- break;
- }
-
- if (!path->nodes[level]) {
- ret = 1;
- goto out;
- }
-
- path->slots[level]++;
- b = path->nodes[level];
-
- while (b) {
- level = btrfs_header_level(b);
-
- if (!should_cow_block(trans, root, b))
- goto cow_done;
-
- btrfs_set_path_blocking(path);
- ret = btrfs_cow_block(trans, root, b,
- path->nodes[level + 1],
- path->slots[level + 1], &b);
- if (ret)
- goto out;
-cow_done:
- path->nodes[level] = b;
- btrfs_clear_path_blocking(path, NULL, 0);
- if (level != 0) {
- ret = setup_nodes_for_search(trans, root, path, b,
- level, ins_len,
- &write_lock_level);
- if (ret == -EAGAIN)
- goto search;
- if (ret)
- goto out;
-
- b = path->nodes[level];
- slot = path->slots[level];
-
- ret = read_block_for_search(trans, root, path,
- &b, level, slot, &key, 0);
- if (ret == -EAGAIN)
- goto search;
- if (ret)
- goto out;
- level = btrfs_header_level(b);
- if (!btrfs_try_tree_write_lock(b)) {
- btrfs_set_path_blocking(path);
- btrfs_tree_lock(b);
- btrfs_clear_path_blocking(path, b,
- BTRFS_WRITE_LOCK);
- }
- path->locks[level] = BTRFS_WRITE_LOCK;
- path->nodes[level] = b;
- path->slots[level] = 0;
- } else {
- path->slots[level] = 0;
- ret = 0;
- break;
- }
- }
-
-out:
- if (ret)
- btrfs_release_path(path);
-
- return ret;
-}
-
int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
u64 time_seq)
{
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 547b7b05727..63c328a9ce9 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -31,10 +31,10 @@
#include <trace/events/btrfs.h>
#include <asm/kmap_types.h>
#include <linux/pagemap.h>
+#include <linux/btrfs.h>
#include "extent_io.h"
#include "extent_map.h"
#include "async-thread.h"
-#include "ioctl.h"
struct btrfs_trans_handle;
struct btrfs_transaction;
@@ -46,7 +46,7 @@ extern struct kmem_cache *btrfs_path_cachep;
extern struct kmem_cache *btrfs_free_space_cachep;
struct btrfs_ordered_sum;
-#define BTRFS_MAGIC "_BHRfS_M"
+#define BTRFS_MAGIC 0x4D5F53665248425FULL /* ascii _BHRfS_M, no null */
#define BTRFS_MAX_MIRRORS 3
@@ -191,6 +191,8 @@ static int btrfs_csum_sizes[] = { 4, 0 };
/* ioprio of readahead is set to idle */
#define BTRFS_IOPRIO_READA (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0))
+#define BTRFS_DIRTY_METADATA_THRESH (32 * 1024 * 1024)
+
/*
* The key defines the order in the tree, and so it also defines (optimal)
* block layout.
@@ -336,7 +338,11 @@ static inline unsigned long btrfs_chunk_item_size(int num_stripes)
/*
* File system states
*/
+#define BTRFS_FS_STATE_ERROR 0
+#define BTRFS_FS_STATE_REMOUNTING 1
+#define BTRFS_FS_STATE_TRANS_ABORTED 2
+/* Super block flags */
/* Errors detected */
#define BTRFS_SUPER_FLAG_ERROR (1ULL << 2)
@@ -502,6 +508,8 @@ struct btrfs_super_block {
#define BTRFS_FEATURE_INCOMPAT_BIG_METADATA (1ULL << 5)
#define BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF (1ULL << 6)
+#define BTRFS_FEATURE_INCOMPAT_RAID56 (1ULL << 7)
+#define BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA (1ULL << 8)
#define BTRFS_FEATURE_COMPAT_SUPP 0ULL
#define BTRFS_FEATURE_COMPAT_RO_SUPP 0ULL
@@ -511,7 +519,9 @@ struct btrfs_super_block {
BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS | \
BTRFS_FEATURE_INCOMPAT_BIG_METADATA | \
BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO | \
- BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF)
+ BTRFS_FEATURE_INCOMPAT_RAID56 | \
+ BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF | \
+ BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
/*
* A leaf is full of items. offset and size tell us where to find
@@ -576,7 +586,6 @@ struct btrfs_path {
unsigned int skip_locking:1;
unsigned int leave_spinning:1;
unsigned int search_commit_root:1;
- unsigned int really_keep_locks:1;
};
/*
@@ -952,8 +961,20 @@ struct btrfs_dev_replace_item {
#define BTRFS_BLOCK_GROUP_RAID1 (1ULL << 4)
#define BTRFS_BLOCK_GROUP_DUP (1ULL << 5)
#define BTRFS_BLOCK_GROUP_RAID10 (1ULL << 6)
+#define BTRFS_BLOCK_GROUP_RAID5 (1 << 7)
+#define BTRFS_BLOCK_GROUP_RAID6 (1 << 8)
#define BTRFS_BLOCK_GROUP_RESERVED BTRFS_AVAIL_ALLOC_BIT_SINGLE
-#define BTRFS_NR_RAID_TYPES 5
+
+enum btrfs_raid_types {
+ BTRFS_RAID_RAID10,
+ BTRFS_RAID_RAID1,
+ BTRFS_RAID_DUP,
+ BTRFS_RAID_RAID0,
+ BTRFS_RAID_SINGLE,
+ BTRFS_RAID_RAID5,
+ BTRFS_RAID_RAID6,
+ BTRFS_NR_RAID_TYPES
+};
#define BTRFS_BLOCK_GROUP_TYPE_MASK (BTRFS_BLOCK_GROUP_DATA | \
BTRFS_BLOCK_GROUP_SYSTEM | \
@@ -961,6 +982,8 @@ struct btrfs_dev_replace_item {
#define BTRFS_BLOCK_GROUP_PROFILE_MASK (BTRFS_BLOCK_GROUP_RAID0 | \
BTRFS_BLOCK_GROUP_RAID1 | \
+ BTRFS_BLOCK_GROUP_RAID5 | \
+ BTRFS_BLOCK_GROUP_RAID6 | \
BTRFS_BLOCK_GROUP_DUP | \
BTRFS_BLOCK_GROUP_RAID10)
/*
@@ -998,9 +1021,9 @@ struct btrfs_block_group_item {
*/
#define BTRFS_QGROUP_STATUS_FLAG_ON (1ULL << 0)
/*
- * SCANNING is set during the initialization phase
+ * RESCAN is set during the initialization phase
*/
-#define BTRFS_QGROUP_STATUS_FLAG_SCANNING (1ULL << 1)
+#define BTRFS_QGROUP_STATUS_FLAG_RESCAN (1ULL << 1)
/*
* Some qgroup entries are known to be out of date,
* either because the configuration has changed in a way that
@@ -1029,7 +1052,7 @@ struct btrfs_qgroup_status_item {
* only used during scanning to record the progress
* of the scan. It contains a logical address
*/
- __le64 scan;
+ __le64 rescan;
} __attribute__ ((__packed__));
struct btrfs_qgroup_info_item {
@@ -1185,6 +1208,10 @@ struct btrfs_block_group_cache {
u64 flags;
u64 sectorsize;
u64 cache_generation;
+
+ /* for raid56, this is a full stripe, without parity */
+ unsigned long full_stripe_len;
+
unsigned int ro:1;
unsigned int dirty:1;
unsigned int iref:1;
@@ -1225,6 +1252,28 @@ struct seq_list {
u64 seq;
};
+enum btrfs_orphan_cleanup_state {
+ ORPHAN_CLEANUP_STARTED = 1,
+ ORPHAN_CLEANUP_DONE = 2,
+};
+
+/* used by the raid56 code to lock stripes for read/modify/write */
+struct btrfs_stripe_hash {
+ struct list_head hash_list;
+ wait_queue_head_t wait;
+ spinlock_t lock;
+};
+
+/* used by the raid56 code to lock stripes for read/modify/write */
+struct btrfs_stripe_hash_table {
+ struct list_head stripe_cache;
+ spinlock_t cache_lock;
+ int cache_size;
+ struct btrfs_stripe_hash table[];
+};
+
+#define BTRFS_STRIPE_HASH_TABLE_BITS 11
+
/* fs_info */
struct reloc_control;
struct btrfs_device;
@@ -1250,6 +1299,7 @@ struct btrfs_fs_info {
/* block group cache stuff */
spinlock_t block_group_cache_lock;
+ u64 first_logical_byte;
struct rb_root block_group_cache_tree;
/* keep track of unallocated space */
@@ -1288,7 +1338,23 @@ struct btrfs_fs_info {
u64 last_trans_log_full_commit;
unsigned long mount_opt;
unsigned long compress_type:4;
+ /*
+ * It is a suggestive number, the read side is safe even it gets a
+ * wrong number because we will write out the data into a regular
+ * extent. The write side(mount/remount) is under ->s_umount lock,
+ * 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;
@@ -1296,6 +1362,17 @@ struct btrfs_fs_info {
wait_queue_head_t transaction_blocked_wait;
wait_queue_head_t async_submit_wait;
+ /*
+ * Used to protect the incompat_flags, compat_flags, compat_ro_flags
+ * when they are updated.
+ *
+ * Because we do not clear the flags for ever, so we needn't use
+ * the lock on the read side.
+ *
+ * We also needn't use the lock when we mount the fs, because
+ * there is no other task which will update the flag.
+ */
+ spinlock_t super_lock;
struct btrfs_super_block *super_copy;
struct btrfs_super_block *super_for_commit;
struct block_device *__bdev;
@@ -1307,6 +1384,13 @@ struct btrfs_fs_info {
struct mutex cleaner_mutex;
struct mutex chunk_mutex;
struct mutex volume_mutex;
+
+ /* this is used during read/modify/write to make sure
+ * no two ios are trying to mod the same stripe at the same
+ * time
+ */
+ struct btrfs_stripe_hash_table *stripe_hash_table;
+
/*
* this protects the ordered operations list only while we are
* processing all of the entries on it. This way we make
@@ -1338,7 +1422,7 @@ struct btrfs_fs_info {
/* this protects tree_mod_seq_list */
spinlock_t tree_mod_seq_lock;
- atomic_t tree_mod_seq;
+ atomic64_t tree_mod_seq;
struct list_head tree_mod_seq_list;
struct seq_list tree_mod_seq_elem;
@@ -1365,6 +1449,7 @@ struct btrfs_fs_info {
*/
struct list_head ordered_extents;
+ spinlock_t delalloc_lock;
/*
* all of the inodes that have delalloc bytes. It is possible for
* this list to be empty even when there is still dirty data=ordered
@@ -1373,13 +1458,6 @@ struct btrfs_fs_info {
struct list_head delalloc_inodes;
/*
- * special rename and truncate targets that must be on disk before
- * we're allowed to commit. This is basically the ext3 style
- * data=ordered list.
- */
- struct list_head ordered_operations;
-
- /*
* there is a pool of worker threads for checksumming during writes
* and a pool for checksumming after reads. This is because readers
* can run with FS locks held, and the writers may be waiting for
@@ -1395,6 +1473,8 @@ struct btrfs_fs_info {
struct btrfs_workers flush_workers;
struct btrfs_workers endio_workers;
struct btrfs_workers endio_meta_workers;
+ struct btrfs_workers endio_raid56_workers;
+ struct btrfs_workers rmw_workers;
struct btrfs_workers endio_meta_write_workers;
struct btrfs_workers endio_write_workers;
struct btrfs_workers endio_freespace_worker;
@@ -1423,10 +1503,12 @@ struct btrfs_fs_info {
u64 total_pinned;
- /* protected by the delalloc lock, used to keep from writing
- * metadata until there is a nice batch
- */
- u64 dirty_metadata_bytes;
+ /* used to keep from writing metadata until there is a nice batch */
+ struct percpu_counter dirty_metadata_bytes;
+ struct percpu_counter delalloc_bytes;
+ s32 dirty_metadata_batch;
+ s32 delalloc_batch;
+
struct list_head dirty_cowonly_roots;
struct btrfs_fs_devices *fs_devices;
@@ -1442,9 +1524,6 @@ struct btrfs_fs_info {
struct reloc_control *reloc_ctl;
- spinlock_t delalloc_lock;
- u64 delalloc_bytes;
-
/* data_alloc_cluster is only used in ssd mode */
struct btrfs_free_cluster data_alloc_cluster;
@@ -1456,6 +1535,8 @@ struct btrfs_fs_info {
struct rb_root defrag_inodes;
atomic_t defrag_running;
+ /* Used to protect avail_{data, metadata, system}_alloc_bits */
+ seqlock_t profiles_lock;
/*
* these three are in extended format (availability of single
* chunks is denoted by BTRFS_AVAIL_ALLOC_BIT_SINGLE bit, other
@@ -1513,14 +1594,22 @@ struct btrfs_fs_info {
struct rb_root qgroup_tree;
spinlock_t qgroup_lock;
+ /* protect user change for quota operations */
+ struct mutex qgroup_ioctl_lock;
+
/* list of dirty qgroups to be written at next commit */
struct list_head dirty_qgroups;
/* used by btrfs_qgroup_record_ref for an efficient tree traversal */
u64 qgroup_seq;
+ /* qgroup rescan items */
+ struct mutex qgroup_rescan_lock; /* protects the progress item */
+ struct btrfs_key qgroup_rescan_progress;
+ struct btrfs_workers qgroup_rescan_workers;
+
/* filesystem state */
- u64 fs_state;
+ unsigned long fs_state;
struct btrfs_delayed_root *delayed_root;
@@ -1623,6 +1712,9 @@ struct btrfs_root {
struct list_head root_list;
+ spinlock_t log_extents_lock[2];
+ struct list_head logged_list[2];
+
spinlock_t orphan_lock;
atomic_t orphan_inodes;
struct btrfs_block_rsv *orphan_block_rsv;
@@ -1737,6 +1829,12 @@ struct btrfs_ioctl_defrag_range_args {
*/
#define BTRFS_EXTENT_ITEM_KEY 168
+/*
+ * The same as the BTRFS_EXTENT_ITEM_KEY, except it's metadata we already know
+ * the length, so we save the level in key->offset instead of the length.
+ */
+#define BTRFS_METADATA_ITEM_KEY 169
+
#define BTRFS_TREE_BLOCK_REF_KEY 176
#define BTRFS_EXTENT_DATA_REF_KEY 178
@@ -1832,6 +1930,7 @@ struct btrfs_ioctl_defrag_range_args {
#define btrfs_clear_opt(o, opt) ((o) &= ~BTRFS_MOUNT_##opt)
#define btrfs_set_opt(o, opt) ((o) |= BTRFS_MOUNT_##opt)
+#define btrfs_raw_test_opt(o, opt) ((o) & BTRFS_MOUNT_##opt)
#define btrfs_test_opt(root, opt) ((root)->fs_info->mount_opt & \
BTRFS_MOUNT_##opt)
/*
@@ -2694,8 +2793,10 @@ BTRFS_SETGET_STACK_FUNCS(super_cache_generation, struct btrfs_super_block,
static inline int btrfs_super_csum_size(struct btrfs_super_block *s)
{
- int t = btrfs_super_csum_type(s);
- BUG_ON(t >= ARRAY_SIZE(btrfs_csum_sizes));
+ u16 t = btrfs_super_csum_type(s);
+ /*
+ * csum type is validated at mount time
+ */
return btrfs_csum_sizes[t];
}
@@ -2792,8 +2893,8 @@ BTRFS_SETGET_FUNCS(qgroup_status_version, struct btrfs_qgroup_status_item,
version, 64);
BTRFS_SETGET_FUNCS(qgroup_status_flags, struct btrfs_qgroup_status_item,
flags, 64);
-BTRFS_SETGET_FUNCS(qgroup_status_scan, struct btrfs_qgroup_status_item,
- scan, 64);
+BTRFS_SETGET_FUNCS(qgroup_status_rescan, struct btrfs_qgroup_status_item,
+ rescan, 64);
/* btrfs_qgroup_info_item */
BTRFS_SETGET_FUNCS(qgroup_info_generation, struct btrfs_qgroup_info_item,
@@ -2933,11 +3034,10 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len);
int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 bytenr,
- u64 num_bytes, u64 *refs, u64 *flags);
+ u64 offset, int metadata, u64 *refs, u64 *flags);
int btrfs_pin_extent(struct btrfs_root *root,
u64 bytenr, u64 num, int reserved);
-int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+int btrfs_pin_extent_for_log_replay(struct btrfs_root *root,
u64 bytenr, u64 num_bytes);
int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -2946,8 +3046,6 @@ struct btrfs_block_group_cache *btrfs_lookup_block_group(
struct btrfs_fs_info *info,
u64 bytenr);
void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
-u64 btrfs_find_block_group(struct btrfs_root *root,
- u64 search_start, u64 search_hint, int owner);
struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u32 blocksize,
u64 parent, u64 root_objectid,
@@ -2957,10 +3055,6 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct extent_buffer *buf,
u64 parent, int last_ref);
-struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- u64 bytenr, u32 blocksize,
- int level);
int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 root_objectid, u64 owner,
@@ -2973,7 +3067,7 @@ int btrfs_reserve_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 num_bytes, u64 min_alloc_size,
u64 empty_size, u64 hint_byte,
- struct btrfs_key *ins, u64 data);
+ struct btrfs_key *ins, int is_data);
int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
struct extent_buffer *buf, int full_backref, int for_cow);
int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
@@ -3013,7 +3107,6 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 group_start);
void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
-u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags);
u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data);
void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
@@ -3035,8 +3128,13 @@ void btrfs_trans_release_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);
-int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending);
+int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
+ struct btrfs_block_rsv *rsv,
+ int nitems,
+ u64 *qgroup_reserved);
+void btrfs_subvolume_release_metadata(struct btrfs_root *root,
+ 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 num_bytes);
@@ -3085,17 +3183,16 @@ int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2);
int btrfs_previous_item(struct btrfs_root *root,
struct btrfs_path *path, u64 min_objectid,
int type);
-void btrfs_set_item_key_safe(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void btrfs_set_item_key_safe(struct btrfs_root *root, struct btrfs_path *path,
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,
struct btrfs_key *key, int lowest_level,
- int cache_only, u64 min_trans);
+ u64 min_trans);
int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
struct btrfs_key *max_key,
- struct btrfs_path *path, int cache_only,
+ struct btrfs_path *path,
u64 min_trans);
enum btrfs_compare_tree_result {
BTRFS_COMPARE_TREE_NEW,
@@ -3122,12 +3219,9 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans,
struct extent_buffer **cow_ret, u64 new_root_objectid);
int btrfs_block_can_be_shared(struct btrfs_root *root,
struct extent_buffer *buf);
-void btrfs_extend_item(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void btrfs_extend_item(struct btrfs_root *root, struct btrfs_path *path,
u32 data_size);
-void btrfs_truncate_item(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
+void btrfs_truncate_item(struct btrfs_root *root, struct btrfs_path *path,
u32 new_size, int from_end);
int btrfs_split_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -3148,7 +3242,7 @@ int btrfs_search_slot_for_read(struct btrfs_root *root,
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, int cache_only, u64 *last_ret,
+ int start_slot, u64 *last_ret,
struct btrfs_key *progress);
void btrfs_release_path(struct btrfs_path *p);
struct btrfs_path *btrfs_alloc_path(void);
@@ -3167,8 +3261,7 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
return btrfs_del_items(trans, root, path, path->slots[0], 1);
}
-void setup_items_for_insert(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
+void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
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
@@ -3188,9 +3281,6 @@ static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
}
int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path);
-int btrfs_next_leaf_write(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *path,
- int del);
int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
u64 time_seq);
static inline int btrfs_next_old_item(struct btrfs_root *root,
@@ -3205,7 +3295,6 @@ static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p)
{
return btrfs_next_old_item(root, p, 0);
}
-int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path);
int btrfs_leaf_free_space(struct btrfs_root *root, struct extent_buffer *leaf);
int __must_check btrfs_drop_snapshot(struct btrfs_root *root,
struct btrfs_block_rsv *block_rsv,
@@ -3242,10 +3331,7 @@ u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
struct seq_list *elem);
void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
struct seq_list *elem);
-static inline u64 btrfs_inc_tree_mod_seq(struct btrfs_fs_info *fs_info)
-{
- return atomic_inc_return(&fs_info->tree_mod_seq);
-}
+u64 btrfs_tree_mod_seq_prev(u64 seq);
int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq);
/* root-item.c */
@@ -3269,9 +3355,8 @@ int __must_check btrfs_update_root(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_key *key,
struct btrfs_root_item *item);
-void btrfs_read_root_item(struct btrfs_root *root,
- struct extent_buffer *eb, int slot,
- struct btrfs_root_item *item);
+void btrfs_read_root_item(struct extent_buffer *eb, int slot,
+ struct btrfs_root_item *item);
int btrfs_find_last_root(struct btrfs_root *root, u64 objectid, struct
btrfs_root_item *item, struct btrfs_key *key);
int btrfs_find_dead_roots(struct btrfs_root *root, u64 objectid);
@@ -3304,9 +3389,6 @@ struct btrfs_dir_item *
btrfs_search_dir_index_item(struct btrfs_root *root,
struct btrfs_path *path, u64 dirid,
const char *name, int name_len);
-struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_root *root,
- struct btrfs_path *path,
- const char *name, int name_len);
int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
@@ -3384,16 +3466,11 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path, u64 objectid,
u64 bytenr, int mod);
-u64 btrfs_file_extent_length(struct btrfs_path *path);
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 btrfs_root *root, struct inode *inode,
struct bio *bio, u64 file_start, int contig);
-struct btrfs_csum_item *btrfs_lookup_csum(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
- u64 bytenr, int cow);
int btrfs_csum_truncate(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct btrfs_path *path,
u64 isize);
@@ -3455,18 +3532,15 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
struct extent_state **cached_state);
-int btrfs_writepages(struct address_space *mapping,
- struct writeback_control *wbc);
int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
struct btrfs_root *new_root, u64 new_dirid);
-int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
- size_t size, struct bio *bio, unsigned long bio_flags);
-
+int btrfs_merge_bio_hook(int rw, 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);
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);
-int btrfs_dirty_inode(struct inode *inode);
struct inode *btrfs_alloc_inode(struct super_block *sb);
void btrfs_destroy_inode(struct inode *inode);
int btrfs_drop_inode(struct inode *inode);
@@ -3484,7 +3558,6 @@ int btrfs_update_inode(struct btrfs_trans_handle *trans,
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_del(struct btrfs_trans_handle *trans, struct inode *inode);
int btrfs_orphan_cleanup(struct btrfs_root *root);
void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
@@ -3535,7 +3608,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
struct inode *inode, u64 start, u64 end);
int btrfs_release_file(struct inode *inode, struct file *file);
-void btrfs_drop_pages(struct page **pages, size_t num_pages);
int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
struct page **pages, size_t num_pages,
loff_t pos, size_t write_bytes,
@@ -3543,7 +3615,7 @@ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
/* tree-defrag.c */
int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, int cache_only);
+ struct btrfs_root *root);
/* sysfs.c */
int btrfs_init_sysfs(void);
@@ -3558,14 +3630,31 @@ int btrfs_sync_fs(struct super_block *sb, int wait);
#ifdef CONFIG_PRINTK
__printf(2, 3)
-void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...);
+void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...);
#else
static inline __printf(2, 3)
-void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...)
+void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
{
}
#endif
+#define btrfs_emerg(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_EMERG fmt, ##args)
+#define btrfs_alert(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_ALERT fmt, ##args)
+#define btrfs_crit(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_CRIT fmt, ##args)
+#define btrfs_err(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_ERR fmt, ##args)
+#define btrfs_warn(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_WARNING fmt, ##args)
+#define btrfs_notice(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_NOTICE fmt, ##args)
+#define btrfs_info(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_INFO fmt, ##args)
+#define btrfs_debug(fs_info, fmt, args...) \
+ btrfs_printk(fs_info, KERN_DEBUG fmt, ##args)
+
__printf(5, 6)
void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
unsigned int line, int errno, const char *fmt, ...);
@@ -3587,11 +3676,28 @@ static inline void __btrfs_set_fs_incompat(struct btrfs_fs_info *fs_info,
disk_super = fs_info->super_copy;
features = btrfs_super_incompat_flags(disk_super);
if (!(features & flag)) {
- features |= flag;
- btrfs_set_super_incompat_flags(disk_super, features);
+ spin_lock(&fs_info->super_lock);
+ features = btrfs_super_incompat_flags(disk_super);
+ if (!(features & flag)) {
+ features |= flag;
+ btrfs_set_super_incompat_flags(disk_super, features);
+ printk(KERN_INFO "btrfs: setting %llu feature flag\n",
+ flag);
+ }
+ spin_unlock(&fs_info->super_lock);
}
}
+#define btrfs_fs_incompat(fs_info, opt) \
+ __btrfs_fs_incompat((fs_info), BTRFS_FEATURE_INCOMPAT_##opt)
+
+static inline int __btrfs_fs_incompat(struct btrfs_fs_info *fs_info, u64 flag)
+{
+ struct btrfs_super_block *disk_super;
+ disk_super = fs_info->super_copy;
+ return !!(btrfs_super_incompat_flags(disk_super) & flag);
+}
+
/*
* Call btrfs_abort_transaction as early as possible when an error condition is
* detected, that way the exact line number is reported.
@@ -3620,11 +3726,14 @@ __printf(5, 6)
void __btrfs_panic(struct btrfs_fs_info *fs_info, const char *function,
unsigned int line, int errno, const char *fmt, ...);
+/*
+ * If BTRFS_MOUNT_PANIC_ON_FATAL_ERROR is in mount_opt, __btrfs_panic
+ * will panic(). Otherwise we BUG() here.
+ */
#define btrfs_panic(fs_info, errno, fmt, args...) \
do { \
- struct btrfs_fs_info *_i = (fs_info); \
- __btrfs_panic(_i, __func__, __LINE__, errno, fmt, ##args); \
- BUG_ON(!(_i->mount_opt & BTRFS_MOUNT_PANIC_ON_FATAL_ERROR)); \
+ __btrfs_panic(fs_info, __func__, __LINE__, errno, fmt, ##args); \
+ BUG(); \
} while (0)
/* acl.c */
@@ -3674,7 +3783,6 @@ void btrfs_scrub_continue_super(struct btrfs_root *root);
int btrfs_scrub_cancel(struct btrfs_fs_info *info);
int btrfs_scrub_cancel_dev(struct btrfs_fs_info *info,
struct btrfs_device *dev);
-int btrfs_scrub_cancel_devid(struct btrfs_root *root, u64 devid);
int btrfs_scrub_progress(struct btrfs_root *root, u64 devid,
struct btrfs_scrub_progress *progress);
@@ -3705,7 +3813,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
int btrfs_quota_disable(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
-int btrfs_quota_rescan(struct btrfs_fs_info *fs_info);
+int btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info);
int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 src, u64 dst);
int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
@@ -3745,4 +3853,11 @@ static inline int is_fstree(u64 rootid)
return 1;
return 0;
}
+
+static inline int btrfs_defrag_cancelled(struct btrfs_fs_info *fs_info)
+{
+ return signal_pending(current);
+}
+
+
#endif
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 34836036f01..f26f38ccd19 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -22,8 +22,9 @@
#include "disk-io.h"
#include "transaction.h"
-#define BTRFS_DELAYED_WRITEBACK 400
-#define BTRFS_DELAYED_BACKGROUND 100
+#define BTRFS_DELAYED_WRITEBACK 512
+#define BTRFS_DELAYED_BACKGROUND 128
+#define BTRFS_DELAYED_BATCH 16
static struct kmem_cache *delayed_node_cache;
@@ -201,7 +202,7 @@ static void btrfs_dequeue_delayed_node(struct btrfs_delayed_root *root,
spin_unlock(&root->lock);
}
-struct btrfs_delayed_node *btrfs_first_delayed_node(
+static struct btrfs_delayed_node *btrfs_first_delayed_node(
struct btrfs_delayed_root *delayed_root)
{
struct list_head *p;
@@ -220,7 +221,7 @@ out:
return node;
}
-struct btrfs_delayed_node *btrfs_next_delayed_node(
+static struct btrfs_delayed_node *btrfs_next_delayed_node(
struct btrfs_delayed_node *node)
{
struct btrfs_delayed_root *delayed_root;
@@ -281,7 +282,7 @@ static inline void btrfs_release_delayed_node(struct btrfs_delayed_node *node)
__btrfs_release_delayed_node(node, 0);
}
-struct btrfs_delayed_node *btrfs_first_prepared_delayed_node(
+static struct btrfs_delayed_node *btrfs_first_prepared_delayed_node(
struct btrfs_delayed_root *delayed_root)
{
struct list_head *p;
@@ -307,7 +308,7 @@ static inline void btrfs_release_prepared_delayed_node(
__btrfs_release_delayed_node(node, 1);
}
-struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
+static struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
{
struct btrfs_delayed_item *item;
item = kmalloc(sizeof(*item) + data_len, GFP_NOFS);
@@ -382,7 +383,7 @@ static struct btrfs_delayed_item *__btrfs_lookup_delayed_item(
return NULL;
}
-struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
+static struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
struct btrfs_delayed_node *delayed_node,
struct btrfs_key *key)
{
@@ -393,45 +394,6 @@ struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
return item;
}
-struct btrfs_delayed_item *__btrfs_lookup_delayed_deletion_item(
- struct btrfs_delayed_node *delayed_node,
- struct btrfs_key *key)
-{
- struct btrfs_delayed_item *item;
-
- item = __btrfs_lookup_delayed_item(&delayed_node->del_root, key,
- NULL, NULL);
- return item;
-}
-
-struct btrfs_delayed_item *__btrfs_search_delayed_insertion_item(
- struct btrfs_delayed_node *delayed_node,
- struct btrfs_key *key)
-{
- struct btrfs_delayed_item *item, *next;
-
- item = __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
- NULL, &next);
- if (!item)
- item = next;
-
- return item;
-}
-
-struct btrfs_delayed_item *__btrfs_search_delayed_deletion_item(
- struct btrfs_delayed_node *delayed_node,
- struct btrfs_key *key)
-{
- struct btrfs_delayed_item *item, *next;
-
- item = __btrfs_lookup_delayed_item(&delayed_node->del_root, key,
- NULL, &next);
- if (!item)
- item = next;
-
- return item;
-}
-
static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
struct btrfs_delayed_item *ins,
int action)
@@ -494,6 +456,15 @@ static int __btrfs_add_delayed_deletion_item(struct btrfs_delayed_node *node,
BTRFS_DELAYED_DELETION_ITEM);
}
+static void finish_one_item(struct btrfs_delayed_root *delayed_root)
+{
+ int seq = atomic_inc_return(&delayed_root->items_seq);
+ if ((atomic_dec_return(&delayed_root->items) <
+ BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0) &&
+ waitqueue_active(&delayed_root->wait))
+ wake_up(&delayed_root->wait);
+}
+
static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
{
struct rb_root *root;
@@ -512,10 +483,8 @@ static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
rb_erase(&delayed_item->rb_node, root);
delayed_item->delayed_node->count--;
- if (atomic_dec_return(&delayed_root->items) <
- BTRFS_DELAYED_BACKGROUND &&
- waitqueue_active(&delayed_root->wait))
- wake_up(&delayed_root->wait);
+
+ finish_one_item(delayed_root);
}
static void btrfs_release_delayed_item(struct btrfs_delayed_item *item)
@@ -527,7 +496,7 @@ static void btrfs_release_delayed_item(struct btrfs_delayed_item *item)
}
}
-struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
+static struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
struct btrfs_delayed_node *delayed_node)
{
struct rb_node *p;
@@ -540,7 +509,7 @@ struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
return item;
}
-struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
+static struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
struct btrfs_delayed_node *delayed_node)
{
struct rb_node *p;
@@ -553,7 +522,7 @@ struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
return item;
}
-struct btrfs_delayed_item *__btrfs_next_delayed_item(
+static struct btrfs_delayed_item *__btrfs_next_delayed_item(
struct btrfs_delayed_item *item)
{
struct rb_node *p;
@@ -758,10 +727,9 @@ static void btrfs_delayed_inode_release_metadata(struct btrfs_root *root,
* This helper will insert some continuous items into the same leaf according
* to the free space of the leaf.
*/
-static int btrfs_batch_insert_items(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
- struct btrfs_delayed_item *item)
+static int btrfs_batch_insert_items(struct btrfs_root *root,
+ struct btrfs_path *path,
+ struct btrfs_delayed_item *item)
{
struct btrfs_delayed_item *curr, *next;
int free_space;
@@ -840,7 +808,7 @@ static int btrfs_batch_insert_items(struct btrfs_trans_handle *trans,
btrfs_clear_path_blocking(path, NULL, 0);
/* insert the keys of the items */
- setup_items_for_insert(trans, root, path, keys, data_size,
+ setup_items_for_insert(root, path, keys, data_size,
total_data_size, total_size, nitems);
/* insert the dir index items */
@@ -875,7 +843,6 @@ static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
struct btrfs_delayed_item *delayed_item)
{
struct extent_buffer *leaf;
- struct btrfs_item *item;
char *ptr;
int ret;
@@ -886,7 +853,6 @@ static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
leaf = path->nodes[0];
- item = btrfs_item_nr(leaf, path->slots[0]);
ptr = btrfs_item_ptr(leaf, path->slots[0], char);
write_extent_buffer(leaf, delayed_item->data, (unsigned long)ptr,
@@ -926,7 +892,7 @@ do_again:
if (curr && btrfs_is_continuous_delayed_item(prev, curr)) {
/* insert the continuous items into the same leaf */
path->slots[0]++;
- btrfs_batch_insert_items(trans, root, path, curr);
+ btrfs_batch_insert_items(root, path, curr);
}
btrfs_release_delayed_item(prev);
btrfs_mark_buffer_dirty(path->nodes[0]);
@@ -1058,39 +1024,29 @@ static void btrfs_release_delayed_inode(struct btrfs_delayed_node *delayed_node)
delayed_node->count--;
delayed_root = delayed_node->root->fs_info->delayed_root;
- if (atomic_dec_return(&delayed_root->items) <
- BTRFS_DELAYED_BACKGROUND &&
- waitqueue_active(&delayed_root->wait))
- wake_up(&delayed_root->wait);
+ finish_one_item(delayed_root);
}
}
-static int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
- struct btrfs_delayed_node *node)
+static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct btrfs_path *path,
+ struct btrfs_delayed_node *node)
{
struct btrfs_key key;
struct btrfs_inode_item *inode_item;
struct extent_buffer *leaf;
int ret;
- mutex_lock(&node->mutex);
- if (!node->inode_dirty) {
- mutex_unlock(&node->mutex);
- return 0;
- }
-
key.objectid = node->inode_id;
btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);
key.offset = 0;
+
ret = btrfs_lookup_inode(trans, root, path, &key, 1);
if (ret > 0) {
btrfs_release_path(path);
- mutex_unlock(&node->mutex);
return -ENOENT;
} else if (ret < 0) {
- mutex_unlock(&node->mutex);
return ret;
}
@@ -1105,11 +1061,47 @@ static int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
btrfs_delayed_inode_release_metadata(root, node);
btrfs_release_delayed_inode(node);
- mutex_unlock(&node->mutex);
return 0;
}
+static inline int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct btrfs_path *path,
+ struct btrfs_delayed_node *node)
+{
+ int ret;
+
+ mutex_lock(&node->mutex);
+ if (!node->inode_dirty) {
+ mutex_unlock(&node->mutex);
+ return 0;
+ }
+
+ ret = __btrfs_update_delayed_inode(trans, root, path, node);
+ mutex_unlock(&node->mutex);
+ return ret;
+}
+
+static inline int
+__btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
+ struct btrfs_path *path,
+ struct btrfs_delayed_node *node)
+{
+ int ret;
+
+ ret = btrfs_insert_delayed_items(trans, path, node->root, node);
+ if (ret)
+ return ret;
+
+ ret = btrfs_delete_delayed_items(trans, path, node->root, node);
+ if (ret)
+ return ret;
+
+ ret = btrfs_update_delayed_inode(trans, node->root, path, node);
+ return ret;
+}
+
/*
* Called when committing the transaction.
* Returns 0 on success.
@@ -1119,7 +1111,6 @@ static int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
struct btrfs_root *root, int nr)
{
- struct btrfs_root *curr_root = root;
struct btrfs_delayed_root *delayed_root;
struct btrfs_delayed_node *curr_node, *prev_node;
struct btrfs_path *path;
@@ -1142,15 +1133,8 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
curr_node = btrfs_first_delayed_node(delayed_root);
while (curr_node && (!count || (count && nr--))) {
- curr_root = curr_node->root;
- ret = btrfs_insert_delayed_items(trans, path, curr_root,
- curr_node);
- if (!ret)
- ret = btrfs_delete_delayed_items(trans, path,
- curr_root, curr_node);
- if (!ret)
- ret = btrfs_update_delayed_inode(trans, curr_root,
- path, curr_node);
+ ret = __btrfs_commit_inode_delayed_items(trans, path,
+ curr_node);
if (ret) {
btrfs_release_delayed_node(curr_node);
curr_node = NULL;
@@ -1183,51 +1167,93 @@ int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans,
return __btrfs_run_delayed_items(trans, root, nr);
}
-static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
- struct btrfs_delayed_node *node)
+int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
+ struct inode *inode)
{
+ struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
struct btrfs_path *path;
struct btrfs_block_rsv *block_rsv;
int ret;
+ if (!delayed_node)
+ return 0;
+
+ mutex_lock(&delayed_node->mutex);
+ if (!delayed_node->count) {
+ mutex_unlock(&delayed_node->mutex);
+ btrfs_release_delayed_node(delayed_node);
+ return 0;
+ }
+ mutex_unlock(&delayed_node->mutex);
+
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
path->leave_spinning = 1;
block_rsv = trans->block_rsv;
- trans->block_rsv = &node->root->fs_info->delayed_block_rsv;
+ trans->block_rsv = &delayed_node->root->fs_info->delayed_block_rsv;
- ret = btrfs_insert_delayed_items(trans, path, node->root, node);
- if (!ret)
- ret = btrfs_delete_delayed_items(trans, path, node->root, node);
- if (!ret)
- ret = btrfs_update_delayed_inode(trans, node->root, path, node);
- btrfs_free_path(path);
+ ret = __btrfs_commit_inode_delayed_items(trans, path, delayed_node);
+ btrfs_release_delayed_node(delayed_node);
+ btrfs_free_path(path);
trans->block_rsv = block_rsv;
+
return ret;
}
-int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
- struct inode *inode)
+int btrfs_commit_inode_delayed_inode(struct inode *inode)
{
+ struct btrfs_trans_handle *trans;
struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
+ struct btrfs_path *path;
+ struct btrfs_block_rsv *block_rsv;
int ret;
if (!delayed_node)
return 0;
mutex_lock(&delayed_node->mutex);
- if (!delayed_node->count) {
+ if (!delayed_node->inode_dirty) {
mutex_unlock(&delayed_node->mutex);
btrfs_release_delayed_node(delayed_node);
return 0;
}
mutex_unlock(&delayed_node->mutex);
- ret = __btrfs_commit_inode_delayed_items(trans, delayed_node);
+ trans = btrfs_join_transaction(delayed_node->root);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+
+ path = btrfs_alloc_path();
+ if (!path) {
+ ret = -ENOMEM;
+ goto trans_out;
+ }
+ path->leave_spinning = 1;
+
+ block_rsv = trans->block_rsv;
+ trans->block_rsv = &delayed_node->root->fs_info->delayed_block_rsv;
+
+ mutex_lock(&delayed_node->mutex);
+ if (delayed_node->inode_dirty)
+ ret = __btrfs_update_delayed_inode(trans, delayed_node->root,
+ path, delayed_node);
+ else
+ ret = 0;
+ mutex_unlock(&delayed_node->mutex);
+
+ btrfs_free_path(path);
+ trans->block_rsv = block_rsv;
+trans_out:
+ btrfs_end_transaction(trans, delayed_node->root);
+ btrfs_btree_balance_dirty(delayed_node->root);
+out:
btrfs_release_delayed_node(delayed_node);
+
return ret;
}
@@ -1243,48 +1269,49 @@ void btrfs_remove_delayed_node(struct inode *inode)
btrfs_release_delayed_node(delayed_node);
}
-struct btrfs_async_delayed_node {
- struct btrfs_root *root;
- struct btrfs_delayed_node *delayed_node;
+struct btrfs_async_delayed_work {
+ struct btrfs_delayed_root *delayed_root;
+ int nr;
struct btrfs_work work;
};
-static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
+static void btrfs_async_run_delayed_root(struct btrfs_work *work)
{
- struct btrfs_async_delayed_node *async_node;
+ struct btrfs_async_delayed_work *async_work;
+ struct btrfs_delayed_root *delayed_root;
struct btrfs_trans_handle *trans;
struct btrfs_path *path;
struct btrfs_delayed_node *delayed_node = NULL;
struct btrfs_root *root;
struct btrfs_block_rsv *block_rsv;
- int need_requeue = 0;
- int ret;
+ int total_done = 0;
- async_node = container_of(work, struct btrfs_async_delayed_node, work);
+ async_work = container_of(work, struct btrfs_async_delayed_work, work);
+ delayed_root = async_work->delayed_root;
path = btrfs_alloc_path();
if (!path)
goto out;
- path->leave_spinning = 1;
- delayed_node = async_node->delayed_node;
+again:
+ if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND / 2)
+ goto free_path;
+
+ delayed_node = btrfs_first_prepared_delayed_node(delayed_root);
+ if (!delayed_node)
+ goto free_path;
+
+ path->leave_spinning = 1;
root = delayed_node->root;
trans = btrfs_join_transaction(root);
if (IS_ERR(trans))
- goto free_path;
+ goto release_path;
block_rsv = trans->block_rsv;
trans->block_rsv = &root->fs_info->delayed_block_rsv;
- ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
- if (!ret)
- ret = btrfs_delete_delayed_items(trans, path, root,
- delayed_node);
-
- if (!ret)
- btrfs_update_delayed_inode(trans, root, path, delayed_node);
-
+ __btrfs_commit_inode_delayed_items(trans, path, delayed_node);
/*
* Maybe new delayed items have been inserted, so we need requeue
* the work. Besides that, we must dequeue the empty delayed nodes
@@ -1310,57 +1337,47 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
* Task1 will sleep until the transaction is commited.
*/
mutex_lock(&delayed_node->mutex);
- if (delayed_node->count)
- need_requeue = 1;
- else
- btrfs_dequeue_delayed_node(root->fs_info->delayed_root,
- delayed_node);
+ btrfs_dequeue_delayed_node(root->fs_info->delayed_root, delayed_node);
mutex_unlock(&delayed_node->mutex);
trans->block_rsv = block_rsv;
btrfs_end_transaction_dmeta(trans, root);
btrfs_btree_balance_dirty_nodelay(root);
+
+release_path:
+ btrfs_release_path(path);
+ total_done++;
+
+ btrfs_release_prepared_delayed_node(delayed_node);
+ if (async_work->nr == 0 || total_done < async_work->nr)
+ goto again;
+
free_path:
btrfs_free_path(path);
out:
- if (need_requeue)
- btrfs_requeue_work(&async_node->work);
- else {
- btrfs_release_prepared_delayed_node(delayed_node);
- kfree(async_node);
- }
+ wake_up(&delayed_root->wait);
+ kfree(async_work);
}
+
static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
- struct btrfs_root *root, int all)
+ struct btrfs_root *root, int nr)
{
- struct btrfs_async_delayed_node *async_node;
- struct btrfs_delayed_node *curr;
- int count = 0;
+ struct btrfs_async_delayed_work *async_work;
-again:
- curr = btrfs_first_prepared_delayed_node(delayed_root);
- if (!curr)
+ if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND)
return 0;
- async_node = kmalloc(sizeof(*async_node), GFP_NOFS);
- if (!async_node) {
- btrfs_release_prepared_delayed_node(curr);
+ async_work = kmalloc(sizeof(*async_work), GFP_NOFS);
+ if (!async_work)
return -ENOMEM;
- }
-
- async_node->root = root;
- async_node->delayed_node = curr;
-
- async_node->work.func = btrfs_async_run_delayed_node_done;
- async_node->work.flags = 0;
- btrfs_queue_worker(&root->fs_info->delayed_workers, &async_node->work);
- count++;
-
- if (all || count < 4)
- goto again;
+ async_work->delayed_root = delayed_root;
+ async_work->work.func = btrfs_async_run_delayed_root;
+ async_work->work.flags = 0;
+ async_work->nr = nr;
+ btrfs_queue_worker(&root->fs_info->delayed_workers, &async_work->work);
return 0;
}
@@ -1371,30 +1388,55 @@ void btrfs_assert_delayed_root_empty(struct btrfs_root *root)
WARN_ON(btrfs_first_delayed_node(delayed_root));
}
+static int refs_newer(struct btrfs_delayed_root *delayed_root,
+ int seq, int count)
+{
+ int val = atomic_read(&delayed_root->items_seq);
+
+ if (val < seq || val >= seq + count)
+ return 1;
+ return 0;
+}
+
void btrfs_balance_delayed_items(struct btrfs_root *root)
{
struct btrfs_delayed_root *delayed_root;
+ int seq;
delayed_root = btrfs_get_delayed_root(root);
if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND)
return;
+ seq = atomic_read(&delayed_root->items_seq);
+
if (atomic_read(&delayed_root->items) >= BTRFS_DELAYED_WRITEBACK) {
int ret;
- ret = btrfs_wq_run_delayed_node(delayed_root, root, 1);
+ DEFINE_WAIT(__wait);
+
+ ret = btrfs_wq_run_delayed_node(delayed_root, root, 0);
if (ret)
return;
- wait_event_interruptible_timeout(
- delayed_root->wait,
- (atomic_read(&delayed_root->items) <
- BTRFS_DELAYED_BACKGROUND),
- HZ);
- return;
+ while (1) {
+ prepare_to_wait(&delayed_root->wait, &__wait,
+ TASK_INTERRUPTIBLE);
+
+ if (refs_newer(delayed_root, seq,
+ BTRFS_DELAYED_BATCH) ||
+ atomic_read(&delayed_root->items) <
+ BTRFS_DELAYED_BACKGROUND) {
+ break;
+ }
+ if (!signal_pending(current))
+ schedule();
+ else
+ break;
+ }
+ finish_wait(&delayed_root->wait, &__wait);
}
- btrfs_wq_run_delayed_node(delayed_root, root, 0);
+ btrfs_wq_run_delayed_node(delayed_root, root, BTRFS_DELAYED_BATCH);
}
/* Will return 0 or -ENOMEM */
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index 4f808e1baee..1d5c5f7abe3 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -43,6 +43,7 @@ struct btrfs_delayed_root {
*/
struct list_head prepare_list;
atomic_t items; /* for delayed items */
+ atomic_t items_seq; /* for delayed items */
int nodes; /* for delayed nodes */
wait_queue_head_t wait;
};
@@ -86,6 +87,7 @@ static inline void btrfs_init_delayed_root(
struct btrfs_delayed_root *delayed_root)
{
atomic_set(&delayed_root->items, 0);
+ atomic_set(&delayed_root->items_seq, 0);
delayed_root->nodes = 0;
spin_lock_init(&delayed_root->lock);
init_waitqueue_head(&delayed_root->wait);
@@ -117,6 +119,7 @@ int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
/* 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);
int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index ae941177339..c219463fb1f 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -23,6 +23,10 @@
#include "delayed-ref.h"
#include "transaction.h"
+struct kmem_cache *btrfs_delayed_ref_head_cachep;
+struct kmem_cache *btrfs_delayed_tree_ref_cachep;
+struct kmem_cache *btrfs_delayed_data_ref_cachep;
+struct kmem_cache *btrfs_delayed_extent_op_cachep;
/*
* delayed back reference update tracking. For subvolume trees
* we queue up extent allocations and backref maintenance for
@@ -36,16 +40,19 @@
* compare two delayed tree backrefs with same bytenr and type
*/
static int comp_tree_refs(struct btrfs_delayed_tree_ref *ref2,
- struct btrfs_delayed_tree_ref *ref1)
+ struct btrfs_delayed_tree_ref *ref1, int type)
{
- if (ref1->root < ref2->root)
- return -1;
- if (ref1->root > ref2->root)
- return 1;
- if (ref1->parent < ref2->parent)
- return -1;
- if (ref1->parent > ref2->parent)
- return 1;
+ if (type == BTRFS_TREE_BLOCK_REF_KEY) {
+ if (ref1->root < ref2->root)
+ return -1;
+ if (ref1->root > ref2->root)
+ return 1;
+ } else {
+ if (ref1->parent < ref2->parent)
+ return -1;
+ if (ref1->parent > ref2->parent)
+ return 1;
+ }
return 0;
}
@@ -109,7 +116,8 @@ static int comp_entry(struct btrfs_delayed_ref_node *ref2,
if (ref1->type == BTRFS_TREE_BLOCK_REF_KEY ||
ref1->type == BTRFS_SHARED_BLOCK_REF_KEY) {
return comp_tree_refs(btrfs_delayed_node_to_tree_ref(ref2),
- btrfs_delayed_node_to_tree_ref(ref1));
+ btrfs_delayed_node_to_tree_ref(ref1),
+ ref1->type);
} else if (ref1->type == BTRFS_EXTENT_DATA_REF_KEY ||
ref1->type == BTRFS_SHARED_DATA_REF_KEY) {
return comp_data_refs(btrfs_delayed_node_to_data_ref(ref2),
@@ -353,8 +361,10 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
elem = list_first_entry(&fs_info->tree_mod_seq_list,
struct seq_list, list);
if (seq >= elem->seq) {
- pr_debug("holding back delayed_ref %llu, lowest is "
- "%llu (%p)\n", seq, elem->seq, delayed_refs);
+ pr_debug("holding back delayed_ref %#x.%x, lowest is %#x.%x (%p)\n",
+ (u32)(seq >> 32), (u32)seq,
+ (u32)(elem->seq >> 32), (u32)elem->seq,
+ delayed_refs);
ret = 1;
}
}
@@ -422,6 +432,14 @@ again:
return 1;
}
+void btrfs_release_ref_cluster(struct list_head *cluster)
+{
+ struct list_head *pos, *q;
+
+ list_for_each_safe(pos, q, cluster)
+ list_del_init(pos);
+}
+
/*
* helper function to update an extent delayed ref in the
* rbtree. existing and update must both have the same
@@ -511,7 +529,7 @@ update_existing_head_ref(struct btrfs_delayed_ref_node *existing,
ref->extent_op->flags_to_set;
existing_ref->extent_op->update_flags = 1;
}
- kfree(ref->extent_op);
+ btrfs_free_delayed_extent_op(ref->extent_op);
}
}
/*
@@ -592,7 +610,7 @@ static noinline void add_delayed_ref_head(struct btrfs_fs_info *fs_info,
* we've updated the existing ref, free the newly
* allocated ref
*/
- kfree(head_ref);
+ kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
} else {
delayed_refs->num_heads++;
delayed_refs->num_heads_ready++;
@@ -653,7 +671,7 @@ static noinline void add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
* we've updated the existing ref, free the newly
* allocated ref
*/
- kfree(full_ref);
+ kmem_cache_free(btrfs_delayed_tree_ref_cachep, full_ref);
} else {
delayed_refs->num_entries++;
trans->delayed_ref_updates++;
@@ -714,7 +732,7 @@ static noinline void add_delayed_data_ref(struct btrfs_fs_info *fs_info,
* we've updated the existing ref, free the newly
* allocated ref
*/
- kfree(full_ref);
+ kmem_cache_free(btrfs_delayed_data_ref_cachep, full_ref);
} else {
delayed_refs->num_entries++;
trans->delayed_ref_updates++;
@@ -738,13 +756,13 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_root *delayed_refs;
BUG_ON(extent_op && extent_op->is_data);
- ref = kmalloc(sizeof(*ref), GFP_NOFS);
+ ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS);
if (!ref)
return -ENOMEM;
- head_ref = kmalloc(sizeof(*head_ref), GFP_NOFS);
+ head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
if (!head_ref) {
- kfree(ref);
+ kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
return -ENOMEM;
}
@@ -786,13 +804,13 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_root *delayed_refs;
BUG_ON(extent_op && !extent_op->is_data);
- ref = kmalloc(sizeof(*ref), GFP_NOFS);
+ ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS);
if (!ref)
return -ENOMEM;
- head_ref = kmalloc(sizeof(*head_ref), GFP_NOFS);
+ head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
if (!head_ref) {
- kfree(ref);
+ kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
return -ENOMEM;
}
@@ -826,7 +844,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_head *head_ref;
struct btrfs_delayed_ref_root *delayed_refs;
- head_ref = kmalloc(sizeof(*head_ref), GFP_NOFS);
+ head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
if (!head_ref)
return -ENOMEM;
@@ -860,3 +878,51 @@ btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, u64 bytenr)
return btrfs_delayed_node_to_head(ref);
return NULL;
}
+
+void btrfs_delayed_ref_exit(void)
+{
+ if (btrfs_delayed_ref_head_cachep)
+ kmem_cache_destroy(btrfs_delayed_ref_head_cachep);
+ if (btrfs_delayed_tree_ref_cachep)
+ kmem_cache_destroy(btrfs_delayed_tree_ref_cachep);
+ if (btrfs_delayed_data_ref_cachep)
+ kmem_cache_destroy(btrfs_delayed_data_ref_cachep);
+ if (btrfs_delayed_extent_op_cachep)
+ kmem_cache_destroy(btrfs_delayed_extent_op_cachep);
+}
+
+int btrfs_delayed_ref_init(void)
+{
+ btrfs_delayed_ref_head_cachep = kmem_cache_create(
+ "btrfs_delayed_ref_head",
+ sizeof(struct btrfs_delayed_ref_head), 0,
+ SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+ if (!btrfs_delayed_ref_head_cachep)
+ goto fail;
+
+ btrfs_delayed_tree_ref_cachep = kmem_cache_create(
+ "btrfs_delayed_tree_ref",
+ sizeof(struct btrfs_delayed_tree_ref), 0,
+ SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+ if (!btrfs_delayed_tree_ref_cachep)
+ goto fail;
+
+ btrfs_delayed_data_ref_cachep = kmem_cache_create(
+ "btrfs_delayed_data_ref",
+ sizeof(struct btrfs_delayed_data_ref), 0,
+ SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+ if (!btrfs_delayed_data_ref_cachep)
+ goto fail;
+
+ btrfs_delayed_extent_op_cachep = kmem_cache_create(
+ "btrfs_delayed_extent_op",
+ sizeof(struct btrfs_delayed_extent_op), 0,
+ SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+ if (!btrfs_delayed_extent_op_cachep)
+ goto fail;
+
+ return 0;
+fail:
+ btrfs_delayed_ref_exit();
+ return -ENOMEM;
+}
diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index c9d703693df..f75fcaf79ae 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -132,6 +132,15 @@ struct btrfs_delayed_ref_root {
unsigned long num_heads_ready;
/*
+ * bumped when someone is making progress on the delayed
+ * refs, so that other procs know they are just adding to
+ * contention intead of helping
+ */
+ atomic_t procs_running_refs;
+ atomic_t ref_seq;
+ wait_queue_head_t wait;
+
+ /*
* set when the tree is flushing before a transaction commit,
* used by the throttling code to decide if new updates need
* to be run right away
@@ -141,12 +150,47 @@ struct btrfs_delayed_ref_root {
u64 run_delayed_start;
};
+extern struct kmem_cache *btrfs_delayed_ref_head_cachep;
+extern struct kmem_cache *btrfs_delayed_tree_ref_cachep;
+extern struct kmem_cache *btrfs_delayed_data_ref_cachep;
+extern struct kmem_cache *btrfs_delayed_extent_op_cachep;
+
+int btrfs_delayed_ref_init(void);
+void btrfs_delayed_ref_exit(void);
+
+static inline struct btrfs_delayed_extent_op *
+btrfs_alloc_delayed_extent_op(void)
+{
+ return kmem_cache_alloc(btrfs_delayed_extent_op_cachep, GFP_NOFS);
+}
+
+static inline void
+btrfs_free_delayed_extent_op(struct btrfs_delayed_extent_op *op)
+{
+ if (op)
+ kmem_cache_free(btrfs_delayed_extent_op_cachep, 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(ref->in_tree);
- kfree(ref);
+ switch (ref->type) {
+ case BTRFS_TREE_BLOCK_REF_KEY:
+ case BTRFS_SHARED_BLOCK_REF_KEY:
+ kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
+ break;
+ case BTRFS_EXTENT_DATA_REF_KEY:
+ case BTRFS_SHARED_DATA_REF_KEY:
+ kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
+ break;
+ case 0:
+ kmem_cache_free(btrfs_delayed_ref_head_cachep, ref);
+ break;
+ default:
+ BUG();
+ }
}
}
@@ -176,8 +220,14 @@ struct btrfs_delayed_ref_head *
btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, 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)
+{
+ mutex_unlock(&head->mutex);
+}
+
int btrfs_find_ref_cluster(struct btrfs_trans_handle *trans,
struct list_head *cluster, u64 search_start);
+void btrfs_release_ref_cluster(struct list_head *cluster);
int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_root *delayed_refs,
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 66dbc8dbddf..7ba7b3900cb 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -465,7 +465,11 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
* flush all outstanding I/O and inode extent mappings before the
* copy operation is declared as being finished
*/
- btrfs_start_delalloc_inodes(root, 0);
+ ret = btrfs_start_delalloc_inodes(root, 0);
+ if (ret) {
+ mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
+ return ret;
+ }
btrfs_wait_ordered_extents(root, 0);
trans = btrfs_start_transaction(root, 0);
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 502c2158167..79e594e341c 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -21,6 +21,10 @@
#include "hash.h"
#include "transaction.h"
+static struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_root *root,
+ struct btrfs_path *path,
+ const char *name, int name_len);
+
/*
* insert a name into a directory, doing overflow properly if there is a hash
* collision. data_size indicates how big the item inserted should be. On
@@ -49,7 +53,7 @@ static struct btrfs_dir_item *insert_with_overflow(struct btrfs_trans_handle
di = btrfs_match_dir_item_name(root, path, name, name_len);
if (di)
return ERR_PTR(-EEXIST);
- btrfs_extend_item(trans, root, path, data_size);
+ btrfs_extend_item(root, path, data_size);
} else if (ret < 0)
return ERR_PTR(ret);
WARN_ON(ret > 0);
@@ -379,7 +383,7 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
* this walks through all the entries in a dir item and finds one
* for a specific name.
*/
-struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_root *root,
+static struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_root *root,
struct btrfs_path *path,
const char *name, int name_len)
{
@@ -442,8 +446,7 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
start = btrfs_item_ptr_offset(leaf, path->slots[0]);
memmove_extent_buffer(leaf, ptr, ptr + sub_item_len,
item_len - (ptr + sub_item_len - start));
- btrfs_truncate_item(trans, root, path,
- item_len - sub_item_len, 1);
+ btrfs_truncate_item(root, path, item_len - sub_item_len, 1);
}
return ret;
}
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index a8f652dc940..4e9ebe1f182 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -30,6 +30,7 @@
#include <linux/slab.h>
#include <linux/migrate.h>
#include <linux/ratelimit.h>
+#include <linux/uuid.h>
#include <asm/unaligned.h>
#include "compat.h"
#include "ctree.h"
@@ -46,6 +47,7 @@
#include "check-integrity.h"
#include "rcu-string.h"
#include "dev-replace.h"
+#include "raid56.h"
#ifdef CONFIG_X86
#include <asm/cpufeature.h>
@@ -56,17 +58,20 @@ 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 void btrfs_destroy_ordered_operations(struct btrfs_root *root);
+static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
+ struct btrfs_root *root);
static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
struct btrfs_root *root);
-static void btrfs_destroy_pending_snapshots(struct btrfs_transaction *t);
+static void btrfs_evict_pending_snapshots(struct btrfs_transaction *t);
static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root);
static int btrfs_destroy_marked_extents(struct btrfs_root *root,
struct extent_io_tree *dirty_pages,
int mark);
static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
struct extent_io_tree *pinned_extents);
+static int btrfs_cleanup_transaction(struct btrfs_root *root);
+static void btrfs_error_commit_super(struct btrfs_root *root);
/*
* end_io_wq structs are used to do processing in task context when an IO is
@@ -220,7 +225,7 @@ static struct extent_map *btree_get_extent(struct inode *inode,
em->bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
+ ret = add_extent_mapping(em_tree, em, 0);
if (ret == -EEXIST) {
free_extent_map(em);
em = lookup_extent_mapping(em_tree, start, len);
@@ -236,7 +241,7 @@ out:
return em;
}
-u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len)
+u32 btrfs_csum_data(char *data, u32 seed, size_t len)
{
return crc32c(seed, data, len);
}
@@ -272,7 +277,7 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
if (err)
return 1;
cur_len = min(len, map_len - (offset - map_start));
- crc = btrfs_csum_data(root, kaddr + offset - map_start,
+ crc = btrfs_csum_data(kaddr + offset - map_start,
crc, cur_len);
len -= cur_len;
offset += cur_len;
@@ -352,6 +357,49 @@ out:
}
/*
+ * Return 0 if the superblock checksum type matches the checksum value of that
+ * algorithm. Pass the raw disk superblock data.
+ */
+static int btrfs_check_super_csum(char *raw_disk_sb)
+{
+ struct btrfs_super_block *disk_sb =
+ (struct btrfs_super_block *)raw_disk_sb;
+ u16 csum_type = btrfs_super_csum_type(disk_sb);
+ int ret = 0;
+
+ if (csum_type == BTRFS_CSUM_TYPE_CRC32) {
+ u32 crc = ~(u32)0;
+ const int csum_size = sizeof(crc);
+ char result[csum_size];
+
+ /*
+ * The super_block structure does not span the whole
+ * BTRFS_SUPER_INFO_SIZE range, we expect that the unused space
+ * is filled with zeros and is included in the checkum.
+ */
+ crc = btrfs_csum_data(raw_disk_sb + BTRFS_CSUM_SIZE,
+ crc, BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);
+ btrfs_csum_final(crc, result);
+
+ if (memcmp(raw_disk_sb, result, csum_size))
+ ret = 1;
+
+ if (ret && btrfs_super_generation(disk_sb) < 10) {
+ printk(KERN_WARNING "btrfs: super block crcs don't match, older mkfs detected\n");
+ ret = 0;
+ }
+ }
+
+ if (csum_type >= ARRAY_SIZE(btrfs_csum_sizes)) {
+ printk(KERN_ERR "btrfs: unsupported checksum algorithm %u\n",
+ csum_type);
+ ret = 1;
+ }
+
+ return ret;
+}
+
+/*
* helper to read a given tree block, doing retries as required when
* the checksums don't match and we have alternate mirrors to try.
*/
@@ -420,7 +468,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_root *root,
static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
{
struct extent_io_tree *tree;
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 found_start;
struct extent_buffer *eb;
@@ -528,41 +576,6 @@ static noinline int check_leaf(struct btrfs_root *root,
return 0;
}
-struct extent_buffer *find_eb_for_page(struct extent_io_tree *tree,
- struct page *page, int max_walk)
-{
- struct extent_buffer *eb;
- u64 start = page_offset(page);
- u64 target = start;
- u64 min_start;
-
- if (start < max_walk)
- min_start = 0;
- else
- min_start = start - max_walk;
-
- while (start >= min_start) {
- eb = find_extent_buffer(tree, start, 0);
- if (eb) {
- /*
- * we found an extent buffer and it contains our page
- * horray!
- */
- if (eb->start <= target &&
- eb->start + eb->len > target)
- return eb;
-
- /* we found an extent buffer that wasn't for us */
- free_extent_buffer(eb);
- return NULL;
- }
- if (start == 0)
- break;
- start -= PAGE_CACHE_SIZE;
- }
- return NULL;
-}
-
static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state, int mirror)
{
@@ -611,6 +624,12 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
goto err;
}
found_level = btrfs_header_level(eb);
+ if (found_level >= BTRFS_MAX_LEVEL) {
+ btrfs_info(root->fs_info, "bad tree block level %d\n",
+ (int)btrfs_header_level(eb));
+ ret = -EIO;
+ goto err;
+ }
btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb),
eb, found_level);
@@ -634,13 +653,19 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
if (!ret)
set_extent_buffer_uptodate(eb);
err:
- if (test_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags)) {
- clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags);
+ if (reads_done &&
+ test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
btree_readahead_hook(root, eb, eb->start, ret);
- }
- if (ret)
+ if (ret) {
+ /*
+ * our io error hook is going to dec the io pages
+ * again, we have to make sure it has something
+ * to decrement
+ */
+ atomic_inc(&eb->io_pages);
clear_extent_buffer_uptodate(eb);
+ }
free_extent_buffer(eb);
out:
return ret;
@@ -654,6 +679,7 @@ static int btree_io_failed_hook(struct page *page, int failed_mirror)
eb = (struct extent_buffer *)page->private;
set_bit(EXTENT_BUFFER_IOERR, &eb->bflags);
eb->read_mirror = failed_mirror;
+ atomic_dec(&eb->io_pages);
if (test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
btree_readahead_hook(root, eb, eb->start, -EIO);
return -EIO; /* we fixed nothing */
@@ -670,17 +696,23 @@ static void end_workqueue_bio(struct bio *bio, int err)
end_io_wq->work.flags = 0;
if (bio->bi_rw & REQ_WRITE) {
- if (end_io_wq->metadata == 1)
+ if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
btrfs_queue_worker(&fs_info->endio_meta_write_workers,
&end_io_wq->work);
- else if (end_io_wq->metadata == 2)
+ else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
btrfs_queue_worker(&fs_info->endio_freespace_worker,
&end_io_wq->work);
+ else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
+ btrfs_queue_worker(&fs_info->endio_raid56_workers,
+ &end_io_wq->work);
else
btrfs_queue_worker(&fs_info->endio_write_workers,
&end_io_wq->work);
} else {
- if (end_io_wq->metadata)
+ if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
+ btrfs_queue_worker(&fs_info->endio_raid56_workers,
+ &end_io_wq->work);
+ else if (end_io_wq->metadata)
btrfs_queue_worker(&fs_info->endio_meta_workers,
&end_io_wq->work);
else
@@ -695,6 +727,7 @@ static void end_workqueue_bio(struct bio *bio, int err)
* 0 - if data
* 1 - if normal metadta
* 2 - if writing to the free space cache area
+ * 3 - raid parity work
*/
int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
int metadata)
@@ -946,18 +979,20 @@ static int btree_writepages(struct address_space *mapping,
struct writeback_control *wbc)
{
struct extent_io_tree *tree;
+ struct btrfs_fs_info *fs_info;
+ int ret;
+
tree = &BTRFS_I(mapping->host)->io_tree;
if (wbc->sync_mode == WB_SYNC_NONE) {
- struct btrfs_root *root = BTRFS_I(mapping->host)->root;
- u64 num_dirty;
- unsigned long thresh = 32 * 1024 * 1024;
if (wbc->for_kupdate)
return 0;
+ fs_info = BTRFS_I(mapping->host)->root->fs_info;
/* this is a bit racy, but that's ok */
- num_dirty = root->fs_info->dirty_metadata_bytes;
- if (num_dirty < thresh)
+ ret = percpu_counter_compare(&fs_info->dirty_metadata_bytes,
+ BTRFS_DIRTY_METADATA_THRESH);
+ if (ret < 0)
return 0;
}
return btree_write_cache_pages(mapping, wbc);
@@ -974,14 +1009,8 @@ static int btree_releasepage(struct page *page, gfp_t gfp_flags)
{
if (PageWriteback(page) || PageDirty(page))
return 0;
- /*
- * We need to mask out eg. __GFP_HIGHMEM and __GFP_DMA32 as we're doing
- * slab allocation from alloc_extent_state down the callchain where
- * it'd hit a BUG_ON as those flags are not allowed.
- */
- gfp_flags &= ~GFP_SLAB_BUG_MASK;
- return try_release_extent_buffer(page, gfp_flags);
+ return try_release_extent_buffer(page);
}
static void btree_invalidatepage(struct page *page, unsigned long offset)
@@ -1125,24 +1154,16 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
void clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
struct extent_buffer *buf)
{
+ struct btrfs_fs_info *fs_info = root->fs_info;
+
if (btrfs_header_generation(buf) ==
- root->fs_info->running_transaction->transid) {
+ fs_info->running_transaction->transid) {
btrfs_assert_tree_locked(buf);
if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)) {
- spin_lock(&root->fs_info->delalloc_lock);
- if (root->fs_info->dirty_metadata_bytes >= buf->len)
- root->fs_info->dirty_metadata_bytes -= buf->len;
- else {
- spin_unlock(&root->fs_info->delalloc_lock);
- btrfs_panic(root->fs_info, -EOVERFLOW,
- "Can't clear %lu bytes from "
- " dirty_mdatadata_bytes (%llu)",
- buf->len,
- root->fs_info->dirty_metadata_bytes);
- }
- spin_unlock(&root->fs_info->delalloc_lock);
-
+ __percpu_counter_add(&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);
@@ -1178,9 +1199,13 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
INIT_LIST_HEAD(&root->dirty_list);
INIT_LIST_HEAD(&root->root_list);
+ INIT_LIST_HEAD(&root->logged_list[0]);
+ INIT_LIST_HEAD(&root->logged_list[1]);
spin_lock_init(&root->orphan_lock);
spin_lock_init(&root->inode_lock);
spin_lock_init(&root->accounting_lock);
+ spin_lock_init(&root->log_extents_lock[0]);
+ spin_lock_init(&root->log_extents_lock[1]);
mutex_init(&root->objectid_mutex);
mutex_init(&root->log_mutex);
init_waitqueue_head(&root->log_writer_wait);
@@ -1260,6 +1285,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
struct btrfs_key key;
int ret = 0;
u64 bytenr;
+ uuid_le uuid;
root = btrfs_alloc_root(fs_info);
if (!root)
@@ -1276,6 +1302,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
0, objectid, NULL, 0, 0, 0);
if (IS_ERR(leaf)) {
ret = PTR_ERR(leaf);
+ leaf = NULL;
goto fail;
}
@@ -1308,6 +1335,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
btrfs_set_root_used(&root->root_item, leaf->len);
btrfs_set_root_last_snapshot(&root->root_item, 0);
btrfs_set_root_dirid(&root->root_item, 0);
+ uuid_le_gen(&uuid);
+ memcpy(root->root_item.uuid, uuid.b, BTRFS_UUID_SIZE);
root->root_item.drop_level = 0;
key.objectid = objectid;
@@ -1319,11 +1348,16 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
btrfs_tree_unlock(leaf);
+ return root;
+
fail:
- if (ret)
- return ERR_PTR(ret);
+ if (leaf) {
+ btrfs_tree_unlock(leaf);
+ free_extent_buffer(leaf);
+ }
+ kfree(root);
- return root;
+ return ERR_PTR(ret);
}
static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
@@ -1455,7 +1489,7 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
if (ret == 0) {
l = path->nodes[0];
slot = path->slots[0];
- btrfs_read_root_item(tree_root, l, slot, &root->root_item);
+ btrfs_read_root_item(l, slot, &root->root_item);
memcpy(&root->root_key, location, sizeof(*location));
}
btrfs_free_path(path);
@@ -1470,6 +1504,14 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
blocksize, generation);
+ if (!root->node || !extent_buffer_uptodate(root->node)) {
+ ret = (!root->node) ? -ENOMEM : -EIO;
+
+ free_extent_buffer(root->node);
+ kfree(root);
+ return ERR_PTR(ret);
+ }
+
root->commit_root = btrfs_root_node(root);
BUG_ON(!root->node); /* -ENOMEM */
out:
@@ -1637,15 +1679,20 @@ static int cleaner_kthread(void *arg)
struct btrfs_root *root = arg;
do {
+ int again = 0;
+
if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
- mutex_trylock(&root->fs_info->cleaner_mutex)) {
- btrfs_run_delayed_iputs(root);
- btrfs_clean_old_snapshots(root);
- mutex_unlock(&root->fs_info->cleaner_mutex);
+ down_read_trylock(&root->fs_info->sb->s_umount)) {
+ if (mutex_trylock(&root->fs_info->cleaner_mutex)) {
+ btrfs_run_delayed_iputs(root);
+ again = btrfs_clean_one_deleted_snapshot(root);
+ mutex_unlock(&root->fs_info->cleaner_mutex);
+ }
btrfs_run_defrag_inodes(root->fs_info);
+ up_read(&root->fs_info->sb->s_umount);
}
- if (!try_to_freeze()) {
+ if (!try_to_freeze() && !again) {
set_current_state(TASK_INTERRUPTIBLE);
if (!kthread_should_stop())
schedule();
@@ -1914,6 +1961,28 @@ static noinline int next_root_backup(struct btrfs_fs_info *info,
return 0;
}
+/* helper to cleanup workers */
+static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
+{
+ btrfs_stop_workers(&fs_info->generic_worker);
+ btrfs_stop_workers(&fs_info->fixup_workers);
+ btrfs_stop_workers(&fs_info->delalloc_workers);
+ btrfs_stop_workers(&fs_info->workers);
+ btrfs_stop_workers(&fs_info->endio_workers);
+ btrfs_stop_workers(&fs_info->endio_meta_workers);
+ btrfs_stop_workers(&fs_info->endio_raid56_workers);
+ btrfs_stop_workers(&fs_info->rmw_workers);
+ btrfs_stop_workers(&fs_info->endio_meta_write_workers);
+ btrfs_stop_workers(&fs_info->endio_write_workers);
+ btrfs_stop_workers(&fs_info->endio_freespace_worker);
+ btrfs_stop_workers(&fs_info->submit_workers);
+ btrfs_stop_workers(&fs_info->delayed_workers);
+ btrfs_stop_workers(&fs_info->caching_workers);
+ btrfs_stop_workers(&fs_info->readahead_workers);
+ btrfs_stop_workers(&fs_info->flush_workers);
+ btrfs_stop_workers(&fs_info->qgroup_rescan_workers);
+}
+
/* helper to cleanup tree roots */
static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
{
@@ -1951,6 +2020,36 @@ static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
}
}
+static void del_fs_roots(struct btrfs_fs_info *fs_info)
+{
+ int ret;
+ struct btrfs_root *gang[8];
+ int i;
+
+ while (!list_empty(&fs_info->dead_roots)) {
+ gang[0] = list_entry(fs_info->dead_roots.next,
+ struct btrfs_root, root_list);
+ list_del(&gang[0]->root_list);
+
+ if (gang[0]->in_radix) {
+ btrfs_free_fs_root(fs_info, gang[0]);
+ } else {
+ free_extent_buffer(gang[0]->node);
+ free_extent_buffer(gang[0]->commit_root);
+ kfree(gang[0]);
+ }
+ }
+
+ while (1) {
+ ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
+ (void **)gang, 0,
+ ARRAY_SIZE(gang));
+ if (!ret)
+ break;
+ for (i = 0; i < ret; i++)
+ btrfs_free_fs_root(fs_info, gang[i]);
+ }
+}
int open_ctree(struct super_block *sb,
struct btrfs_fs_devices *fs_devices,
@@ -2004,10 +2103,24 @@ int open_ctree(struct super_block *sb,
goto fail_srcu;
}
+ ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0);
+ if (ret) {
+ err = ret;
+ goto fail_bdi;
+ }
+ fs_info->dirty_metadata_batch = PAGE_CACHE_SIZE *
+ (1 + ilog2(nr_cpu_ids));
+
+ ret = percpu_counter_init(&fs_info->delalloc_bytes, 0);
+ if (ret) {
+ err = ret;
+ goto fail_dirty_metadata_bytes;
+ }
+
fs_info->btree_inode = new_inode(sb);
if (!fs_info->btree_inode) {
err = -ENOMEM;
- goto fail_bdi;
+ goto fail_delalloc_bytes;
}
mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS);
@@ -2017,7 +2130,6 @@ int open_ctree(struct super_block *sb,
INIT_LIST_HEAD(&fs_info->dead_roots);
INIT_LIST_HEAD(&fs_info->delayed_iputs);
INIT_LIST_HEAD(&fs_info->delalloc_inodes);
- INIT_LIST_HEAD(&fs_info->ordered_operations);
INIT_LIST_HEAD(&fs_info->caching_block_groups);
spin_lock_init(&fs_info->delalloc_lock);
spin_lock_init(&fs_info->trans_lock);
@@ -2026,8 +2138,10 @@ int open_ctree(struct super_block *sb,
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);
rwlock_init(&fs_info->tree_mod_log_lock);
mutex_init(&fs_info->reloc_mutex);
+ seqlock_init(&fs_info->profiles_lock);
init_completion(&fs_info->kobj_unregister);
INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
@@ -2048,7 +2162,7 @@ int open_ctree(struct super_block *sb,
atomic_set(&fs_info->async_submit_draining, 0);
atomic_set(&fs_info->nr_async_bios, 0);
atomic_set(&fs_info->defrag_running, 0);
- atomic_set(&fs_info->tree_mod_seq, 0);
+ atomic64_set(&fs_info->tree_mod_seq, 0);
fs_info->sb = sb;
fs_info->max_inline = 8192 * 1024;
fs_info->metadata_ratio = 0;
@@ -2126,6 +2240,7 @@ int open_ctree(struct super_block *sb,
spin_lock_init(&fs_info->block_group_cache_lock);
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);
@@ -2151,11 +2266,13 @@ int open_ctree(struct super_block *sb,
mutex_init(&fs_info->dev_replace.lock);
spin_lock_init(&fs_info->qgroup_lock);
+ mutex_init(&fs_info->qgroup_ioctl_lock);
fs_info->qgroup_tree = RB_ROOT;
INIT_LIST_HEAD(&fs_info->dirty_qgroups);
fs_info->qgroup_seq = 1;
fs_info->quota_enabled = 0;
fs_info->pending_quota_state = 0;
+ mutex_init(&fs_info->qgroup_rescan_lock);
btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
btrfs_init_free_cluster(&fs_info->data_alloc_cluster);
@@ -2165,16 +2282,41 @@ int open_ctree(struct super_block *sb,
init_waitqueue_head(&fs_info->transaction_blocked_wait);
init_waitqueue_head(&fs_info->async_submit_wait);
+ ret = btrfs_alloc_stripe_hash_table(fs_info);
+ if (ret) {
+ err = ret;
+ goto fail_alloc;
+ }
+
__setup_root(4096, 4096, 4096, 4096, tree_root,
fs_info, BTRFS_ROOT_TREE_OBJECTID);
invalidate_bdev(fs_devices->latest_bdev);
+
+ /*
+ * Read super block and check the signature bytes only
+ */
bh = btrfs_read_dev_super(fs_devices->latest_bdev);
if (!bh) {
err = -EINVAL;
goto fail_alloc;
}
+ /*
+ * We want to check superblock checksum, the type is stored inside.
+ * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
+ */
+ if (btrfs_check_super_csum(bh->b_data)) {
+ printk(KERN_ERR "btrfs: superblock checksum mismatch\n");
+ err = -EINVAL;
+ goto fail_alloc;
+ }
+
+ /*
+ * super_copy is zeroed at allocation time and we never touch the
+ * following bytes up to INFO_SIZE, the checksum is calculated from
+ * the whole block of INFO_SIZE
+ */
memcpy(fs_info->super_copy, bh->b_data, sizeof(*fs_info->super_copy));
memcpy(fs_info->super_for_commit, fs_info->super_copy,
sizeof(*fs_info->super_for_commit));
@@ -2182,20 +2324,21 @@ int open_ctree(struct super_block *sb,
memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE);
- disk_super = fs_info->super_copy;
- if (!btrfs_super_root(disk_super))
- goto fail_alloc;
-
- /* check FS state, whether FS is broken. */
- fs_info->fs_state |= btrfs_super_flags(disk_super);
-
ret = btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
if (ret) {
printk(KERN_ERR "btrfs: superblock contains fatal errors\n");
- err = ret;
+ err = -EINVAL;
goto fail_alloc;
}
+ disk_super = fs_info->super_copy;
+ if (!btrfs_super_root(disk_super))
+ goto fail_alloc;
+
+ /* check FS state, whether FS is broken. */
+ if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_ERROR)
+ set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
+
/*
* run through our array of backup supers and setup
* our ring pointer to the oldest one
@@ -2247,6 +2390,9 @@ int open_ctree(struct super_block *sb,
if (tree_root->fs_info->compress_type == BTRFS_COMPRESS_LZO)
features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
+ if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
+ printk(KERN_ERR "btrfs: has skinny extents\n");
+
/*
* flag our filesystem as having big metadata blocks if
* they are bigger than the page size
@@ -2261,6 +2407,8 @@ int open_ctree(struct super_block *sb,
leafsize = btrfs_super_leafsize(disk_super);
sectorsize = btrfs_super_sectorsize(disk_super);
stripesize = btrfs_super_stripesize(disk_super);
+ fs_info->dirty_metadata_batch = leafsize * (1 + ilog2(nr_cpu_ids));
+ fs_info->delalloc_batch = sectorsize * 512 * (1 + ilog2(nr_cpu_ids));
/*
* mixed block groups end up with duplicate but slightly offset
@@ -2274,6 +2422,10 @@ int open_ctree(struct super_block *sb,
goto fail_alloc;
}
+ /*
+ * Needn't use the lock because there is no other task which will
+ * update the flag.
+ */
btrfs_set_super_incompat_flags(disk_super, features);
features = btrfs_super_compat_ro_flags(disk_super) &
@@ -2332,6 +2484,12 @@ int open_ctree(struct super_block *sb,
btrfs_init_workers(&fs_info->endio_meta_write_workers,
"endio-meta-write", fs_info->thread_pool_size,
&fs_info->generic_worker);
+ btrfs_init_workers(&fs_info->endio_raid56_workers,
+ "endio-raid56", fs_info->thread_pool_size,
+ &fs_info->generic_worker);
+ btrfs_init_workers(&fs_info->rmw_workers,
+ "rmw", fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->endio_write_workers, "endio-write",
fs_info->thread_pool_size,
&fs_info->generic_worker);
@@ -2343,6 +2501,8 @@ int open_ctree(struct super_block *sb,
btrfs_init_workers(&fs_info->readahead_workers, "readahead",
fs_info->thread_pool_size,
&fs_info->generic_worker);
+ btrfs_init_workers(&fs_info->qgroup_rescan_workers, "qgroup-rescan", 1,
+ &fs_info->generic_worker);
/*
* endios are largely parallel and should have a very
@@ -2350,6 +2510,8 @@ int open_ctree(struct super_block *sb,
*/
fs_info->endio_workers.idle_thresh = 4;
fs_info->endio_meta_workers.idle_thresh = 4;
+ fs_info->endio_raid56_workers.idle_thresh = 4;
+ fs_info->rmw_workers.idle_thresh = 2;
fs_info->endio_write_workers.idle_thresh = 2;
fs_info->endio_meta_write_workers.idle_thresh = 2;
@@ -2366,6 +2528,8 @@ int open_ctree(struct super_block *sb,
ret |= btrfs_start_workers(&fs_info->fixup_workers);
ret |= btrfs_start_workers(&fs_info->endio_workers);
ret |= btrfs_start_workers(&fs_info->endio_meta_workers);
+ ret |= btrfs_start_workers(&fs_info->rmw_workers);
+ ret |= btrfs_start_workers(&fs_info->endio_raid56_workers);
ret |= btrfs_start_workers(&fs_info->endio_meta_write_workers);
ret |= btrfs_start_workers(&fs_info->endio_write_workers);
ret |= btrfs_start_workers(&fs_info->endio_freespace_worker);
@@ -2373,6 +2537,7 @@ int open_ctree(struct super_block *sb,
ret |= btrfs_start_workers(&fs_info->caching_workers);
ret |= btrfs_start_workers(&fs_info->readahead_workers);
ret |= btrfs_start_workers(&fs_info->flush_workers);
+ ret |= btrfs_start_workers(&fs_info->qgroup_rescan_workers);
if (ret) {
err = -ENOMEM;
goto fail_sb_buffer;
@@ -2390,8 +2555,7 @@ int open_ctree(struct super_block *sb,
sb->s_blocksize = sectorsize;
sb->s_blocksize_bits = blksize_bits(sectorsize);
- if (strncmp((char *)(&disk_super->magic), BTRFS_MAGIC,
- sizeof(disk_super->magic))) {
+ if (disk_super->magic != cpu_to_le64(BTRFS_MAGIC)) {
printk(KERN_INFO "btrfs: valid FS not found on %s\n", sb->s_id);
goto fail_sb_buffer;
}
@@ -2421,8 +2585,8 @@ int open_ctree(struct super_block *sb,
chunk_root->node = read_tree_block(chunk_root,
btrfs_super_chunk_root(disk_super),
blocksize, generation);
- BUG_ON(!chunk_root->node); /* -ENOMEM */
- if (!test_bit(EXTENT_BUFFER_UPTODATE, &chunk_root->node->bflags)) {
+ if (!chunk_root->node ||
+ !test_bit(EXTENT_BUFFER_UPTODATE, &chunk_root->node->bflags)) {
printk(KERN_WARNING "btrfs: failed to read chunk root on %s\n",
sb->s_id);
goto fail_tree_roots;
@@ -2607,6 +2771,13 @@ retry_root_backup:
log_tree_root->node = read_tree_block(tree_root, bytenr,
blocksize,
generation + 1);
+ if (!log_tree_root->node ||
+ !extent_buffer_uptodate(log_tree_root->node)) {
+ printk(KERN_ERR "btrfs: failed to read log tree\n");
+ free_extent_buffer(log_tree_root->node);
+ kfree(log_tree_root);
+ goto fail_trans_kthread;
+ }
/* returns with log_tree_root freed on success */
ret = btrfs_recover_log_trees(log_tree_root);
if (ret) {
@@ -2686,6 +2857,8 @@ fail_qgroup:
btrfs_free_qgroup_config(fs_info);
fail_trans_kthread:
kthread_stop(fs_info->transaction_kthread);
+ del_fs_roots(fs_info);
+ btrfs_cleanup_transaction(fs_info->tree_root);
fail_cleaner:
kthread_stop(fs_info->cleaner_kthread);
@@ -2694,40 +2867,32 @@ fail_cleaner:
* kthreads
*/
filemap_write_and_wait(fs_info->btree_inode->i_mapping);
- invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
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);
+ invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
fail_sb_buffer:
- btrfs_stop_workers(&fs_info->generic_worker);
- btrfs_stop_workers(&fs_info->readahead_workers);
- btrfs_stop_workers(&fs_info->fixup_workers);
- btrfs_stop_workers(&fs_info->delalloc_workers);
- btrfs_stop_workers(&fs_info->workers);
- btrfs_stop_workers(&fs_info->endio_workers);
- btrfs_stop_workers(&fs_info->endio_meta_workers);
- btrfs_stop_workers(&fs_info->endio_meta_write_workers);
- btrfs_stop_workers(&fs_info->endio_write_workers);
- btrfs_stop_workers(&fs_info->endio_freespace_worker);
- btrfs_stop_workers(&fs_info->submit_workers);
- btrfs_stop_workers(&fs_info->delayed_workers);
- btrfs_stop_workers(&fs_info->caching_workers);
- btrfs_stop_workers(&fs_info->flush_workers);
+ btrfs_stop_all_workers(fs_info);
fail_alloc:
fail_iput:
btrfs_mapping_tree_free(&fs_info->mapping_tree);
- invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
iput(fs_info->btree_inode);
+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:
+ btrfs_free_stripe_hash_table(fs_info);
btrfs_close_devices(fs_info->fs_devices);
return err;
@@ -2795,8 +2960,7 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev)
super = (struct btrfs_super_block *)bh->b_data;
if (btrfs_super_bytenr(super) != bytenr ||
- strncmp((char *)(&super->magic), BTRFS_MAGIC,
- sizeof(super->magic))) {
+ super->magic != cpu_to_le64(BTRFS_MAGIC)) {
brelse(bh);
continue;
}
@@ -2845,7 +3009,10 @@ static int write_dev_supers(struct btrfs_device *device,
if (wait) {
bh = __find_get_block(device->bdev, bytenr / 4096,
BTRFS_SUPER_INFO_SIZE);
- BUG_ON(!bh);
+ if (!bh) {
+ errors++;
+ continue;
+ }
wait_on_buffer(bh);
if (!buffer_uptodate(bh))
errors++;
@@ -2860,7 +3027,7 @@ static int write_dev_supers(struct btrfs_device *device,
btrfs_set_super_bytenr(sb, bytenr);
crc = ~(u32)0;
- crc = btrfs_csum_data(NULL, (char *)sb +
+ crc = btrfs_csum_data((char *)sb +
BTRFS_CSUM_SIZE, crc,
BTRFS_SUPER_INFO_SIZE -
BTRFS_CSUM_SIZE);
@@ -2872,6 +3039,13 @@ static int write_dev_supers(struct btrfs_device *device,
*/
bh = __getblk(device->bdev, bytenr / 4096,
BTRFS_SUPER_INFO_SIZE);
+ if (!bh) {
+ printk(KERN_ERR "btrfs: couldn't get super "
+ "buffer head for bytenr %Lu\n", bytenr);
+ errors++;
+ continue;
+ }
+
memcpy(bh->b_data, sb, BTRFS_SUPER_INFO_SIZE);
/* one reference for submit_bh */
@@ -3076,11 +3250,16 @@ int btrfs_calc_num_tolerated_disk_barrier_failures(
((flags & BTRFS_BLOCK_GROUP_PROFILE_MASK)
== 0)))
num_tolerated_disk_barrier_failures = 0;
- else if (num_tolerated_disk_barrier_failures > 1
- &&
- (flags & (BTRFS_BLOCK_GROUP_RAID1 |
- BTRFS_BLOCK_GROUP_RAID10)))
- num_tolerated_disk_barrier_failures = 1;
+ else if (num_tolerated_disk_barrier_failures > 1) {
+ if (flags & (BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID10)) {
+ num_tolerated_disk_barrier_failures = 1;
+ } else if (flags &
+ BTRFS_BLOCK_GROUP_RAID5) {
+ num_tolerated_disk_barrier_failures = 2;
+ }
+ }
}
}
up_read(&sinfo->groups_sem);
@@ -3089,7 +3268,7 @@ int btrfs_calc_num_tolerated_disk_barrier_failures(
return num_tolerated_disk_barrier_failures;
}
-int write_all_supers(struct btrfs_root *root, int max_mirrors)
+static int write_all_supers(struct btrfs_root *root, int max_mirrors)
{
struct list_head *head;
struct btrfs_device *dev;
@@ -3195,6 +3374,11 @@ void btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
if (btrfs_root_refs(&root->root_item) == 0)
synchronize_srcu(&fs_info->subvol_srcu);
+ if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
+ btrfs_free_log(NULL, root);
+ btrfs_free_log_root_tree(NULL, fs_info);
+ }
+
__btrfs_remove_free_space_cache(root->free_ino_pinned);
__btrfs_remove_free_space_cache(root->free_ino_ctl);
free_fs_root(root);
@@ -3214,37 +3398,6 @@ static void free_fs_root(struct btrfs_root *root)
kfree(root);
}
-static void del_fs_roots(struct btrfs_fs_info *fs_info)
-{
- int ret;
- struct btrfs_root *gang[8];
- int i;
-
- while (!list_empty(&fs_info->dead_roots)) {
- gang[0] = list_entry(fs_info->dead_roots.next,
- struct btrfs_root, root_list);
- list_del(&gang[0]->root_list);
-
- if (gang[0]->in_radix) {
- btrfs_free_fs_root(fs_info, gang[0]);
- } else {
- free_extent_buffer(gang[0]->node);
- free_extent_buffer(gang[0]->commit_root);
- kfree(gang[0]);
- }
- }
-
- while (1) {
- ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
- (void **)gang, 0,
- ARRAY_SIZE(gang));
- if (!ret)
- break;
- for (i = 0; i < ret; i++)
- btrfs_free_fs_root(fs_info, gang[i]);
- }
-}
-
int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
{
u64 root_objectid = 0;
@@ -3280,8 +3433,8 @@ int btrfs_commit_super(struct btrfs_root *root)
mutex_lock(&root->fs_info->cleaner_mutex);
btrfs_run_delayed_iputs(root);
- btrfs_clean_old_snapshots(root);
mutex_unlock(&root->fs_info->cleaner_mutex);
+ wake_up_process(root->fs_info->cleaner_kthread);
/* wait until ongoing cleanup work done */
down_write(&root->fs_info->cleanup_work_sem);
@@ -3339,7 +3492,7 @@ int close_ctree(struct btrfs_root *root)
printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
}
- if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
+ if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
btrfs_error_commit_super(root);
btrfs_put_block_group_cache(fs_info);
@@ -3352,25 +3505,12 @@ int close_ctree(struct btrfs_root *root)
btrfs_free_qgroup_config(root->fs_info);
- if (fs_info->delalloc_bytes) {
- printk(KERN_INFO "btrfs: at unmount delalloc count %llu\n",
- (unsigned long long)fs_info->delalloc_bytes);
+ if (percpu_counter_sum(&fs_info->delalloc_bytes)) {
+ printk(KERN_INFO "btrfs: at unmount delalloc count %lld\n",
+ percpu_counter_sum(&fs_info->delalloc_bytes));
}
- free_extent_buffer(fs_info->extent_root->node);
- free_extent_buffer(fs_info->extent_root->commit_root);
- free_extent_buffer(fs_info->tree_root->node);
- free_extent_buffer(fs_info->tree_root->commit_root);
- free_extent_buffer(fs_info->chunk_root->node);
- free_extent_buffer(fs_info->chunk_root->commit_root);
- free_extent_buffer(fs_info->dev_root->node);
- free_extent_buffer(fs_info->dev_root->commit_root);
- free_extent_buffer(fs_info->csum_root->node);
- free_extent_buffer(fs_info->csum_root->commit_root);
- if (fs_info->quota_root) {
- free_extent_buffer(fs_info->quota_root->node);
- free_extent_buffer(fs_info->quota_root->commit_root);
- }
+ free_root_pointers(fs_info, 1);
btrfs_free_block_groups(fs_info);
@@ -3378,20 +3518,7 @@ int close_ctree(struct btrfs_root *root)
iput(fs_info->btree_inode);
- btrfs_stop_workers(&fs_info->generic_worker);
- btrfs_stop_workers(&fs_info->fixup_workers);
- btrfs_stop_workers(&fs_info->delalloc_workers);
- btrfs_stop_workers(&fs_info->workers);
- btrfs_stop_workers(&fs_info->endio_workers);
- btrfs_stop_workers(&fs_info->endio_meta_workers);
- btrfs_stop_workers(&fs_info->endio_meta_write_workers);
- btrfs_stop_workers(&fs_info->endio_write_workers);
- btrfs_stop_workers(&fs_info->endio_freespace_worker);
- btrfs_stop_workers(&fs_info->submit_workers);
- btrfs_stop_workers(&fs_info->delayed_workers);
- btrfs_stop_workers(&fs_info->caching_workers);
- btrfs_stop_workers(&fs_info->readahead_workers);
- btrfs_stop_workers(&fs_info->flush_workers);
+ btrfs_stop_all_workers(fs_info);
#ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
if (btrfs_test_opt(root, CHECK_INTEGRITY))
@@ -3401,9 +3528,13 @@ int close_ctree(struct btrfs_root *root)
btrfs_close_devices(fs_info->fs_devices);
btrfs_mapping_tree_free(&fs_info->mapping_tree);
+ percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
+ percpu_counter_destroy(&fs_info->delalloc_bytes);
bdi_destroy(&fs_info->bdi);
cleanup_srcu_struct(&fs_info->subvol_srcu);
+ btrfs_free_stripe_hash_table(fs_info);
+
return 0;
}
@@ -3443,11 +3574,10 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
(unsigned long long)transid,
(unsigned long long)root->fs_info->generation);
was_dirty = set_extent_buffer_dirty(buf);
- if (!was_dirty) {
- spin_lock(&root->fs_info->delalloc_lock);
- root->fs_info->dirty_metadata_bytes += buf->len;
- spin_unlock(&root->fs_info->delalloc_lock);
- }
+ if (!was_dirty)
+ __percpu_counter_add(&root->fs_info->dirty_metadata_bytes,
+ buf->len,
+ root->fs_info->dirty_metadata_batch);
}
static void __btrfs_btree_balance_dirty(struct btrfs_root *root,
@@ -3457,8 +3587,7 @@ static void __btrfs_btree_balance_dirty(struct btrfs_root *root,
* looks as though older kernels can get into trouble with
* this code, they end up stuck in balance_dirty_pages forever
*/
- u64 num_dirty;
- unsigned long thresh = 32 * 1024 * 1024;
+ int ret;
if (current->flags & PF_MEMALLOC)
return;
@@ -3466,9 +3595,9 @@ static void __btrfs_btree_balance_dirty(struct btrfs_root *root,
if (flush_delayed)
btrfs_balance_delayed_items(root);
- num_dirty = root->fs_info->dirty_metadata_bytes;
-
- if (num_dirty > thresh) {
+ ret = percpu_counter_compare(&root->fs_info->dirty_metadata_bytes,
+ BTRFS_DIRTY_METADATA_THRESH);
+ if (ret > 0) {
balance_dirty_pages_ratelimited(
root->fs_info->btree_inode->i_mapping);
}
@@ -3494,18 +3623,13 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
int read_only)
{
- if (btrfs_super_csum_type(fs_info->super_copy) >= ARRAY_SIZE(btrfs_csum_sizes)) {
- printk(KERN_ERR "btrfs: unsupported checksum algorithm\n");
- return -EINVAL;
- }
-
- if (read_only)
- return 0;
-
+ /*
+ * Placeholder for checks
+ */
return 0;
}
-void btrfs_error_commit_super(struct btrfs_root *root)
+static void btrfs_error_commit_super(struct btrfs_root *root)
{
mutex_lock(&root->fs_info->cleaner_mutex);
btrfs_run_delayed_iputs(root);
@@ -3518,7 +3642,8 @@ void btrfs_error_commit_super(struct btrfs_root *root)
btrfs_cleanup_transaction(root);
}
-static void btrfs_destroy_ordered_operations(struct btrfs_root *root)
+static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
+ struct btrfs_root *root)
{
struct btrfs_inode *btrfs_inode;
struct list_head splice;
@@ -3528,7 +3653,7 @@ static void btrfs_destroy_ordered_operations(struct btrfs_root *root)
mutex_lock(&root->fs_info->ordered_operations_mutex);
spin_lock(&root->fs_info->ordered_extent_lock);
- list_splice_init(&root->fs_info->ordered_operations, &splice);
+ list_splice_init(&t->ordered_operations, &splice);
while (!list_empty(&splice)) {
btrfs_inode = list_entry(splice.next, struct btrfs_inode,
ordered_operations);
@@ -3544,35 +3669,16 @@ static void btrfs_destroy_ordered_operations(struct btrfs_root *root)
static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
{
- struct list_head splice;
struct btrfs_ordered_extent *ordered;
- struct inode *inode;
-
- INIT_LIST_HEAD(&splice);
spin_lock(&root->fs_info->ordered_extent_lock);
-
- list_splice_init(&root->fs_info->ordered_extents, &splice);
- while (!list_empty(&splice)) {
- ordered = list_entry(splice.next, struct btrfs_ordered_extent,
- root_extent_list);
-
- list_del_init(&ordered->root_extent_list);
- atomic_inc(&ordered->refs);
-
- /* the inode may be getting freed (in sys_unlink path). */
- inode = igrab(ordered->inode);
-
- spin_unlock(&root->fs_info->ordered_extent_lock);
- if (inode)
- iput(inode);
-
- atomic_set(&ordered->refs, 1);
- btrfs_put_ordered_extent(ordered);
-
- spin_lock(&root->fs_info->ordered_extent_lock);
- }
-
+ /*
+ * This will just short circuit the ordered completion stuff which will
+ * make sure the ordered extent gets properly cleaned up.
+ */
+ list_for_each_entry(ordered, &root->fs_info->ordered_extents,
+ root_extent_list)
+ set_bit(BTRFS_ORDERED_IOERR, &ordered->flags);
spin_unlock(&root->fs_info->ordered_extent_lock);
}
@@ -3594,11 +3700,11 @@ int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
}
while ((node = rb_first(&delayed_refs->root)) != NULL) {
- ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
+ struct btrfs_delayed_ref_head *head = NULL;
+ ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
atomic_set(&ref->refs, 1);
if (btrfs_delayed_ref_is_head(ref)) {
- struct btrfs_delayed_ref_head *head;
head = btrfs_delayed_node_to_head(ref);
if (!mutex_trylock(&head->mutex)) {
@@ -3614,16 +3720,21 @@ int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
continue;
}
- kfree(head->extent_op);
+ if (head->must_insert_reserved)
+ btrfs_pin_extent(root, ref->bytenr,
+ ref->num_bytes, 1);
+ btrfs_free_delayed_extent_op(head->extent_op);
delayed_refs->num_heads--;
if (list_empty(&head->cluster))
delayed_refs->num_heads_ready--;
list_del_init(&head->cluster);
}
+
ref->in_tree = 0;
rb_erase(&ref->rb_node, &delayed_refs->root);
delayed_refs->num_entries--;
-
+ if (head)
+ mutex_unlock(&head->mutex);
spin_unlock(&delayed_refs->lock);
btrfs_put_delayed_ref(ref);
@@ -3636,7 +3747,7 @@ int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
return ret;
}
-static void btrfs_destroy_pending_snapshots(struct btrfs_transaction *t)
+static void btrfs_evict_pending_snapshots(struct btrfs_transaction *t)
{
struct btrfs_pending_snapshot *snapshot;
struct list_head splice;
@@ -3649,10 +3760,8 @@ static void btrfs_destroy_pending_snapshots(struct btrfs_transaction *t)
snapshot = list_entry(splice.next,
struct btrfs_pending_snapshot,
list);
-
+ snapshot->error = -ECANCELED;
list_del_init(&snapshot->list);
-
- kfree(snapshot);
}
}
@@ -3671,6 +3780,8 @@ static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root)
delalloc_inodes);
list_del_init(&btrfs_inode->delalloc_inodes);
+ clear_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &btrfs_inode->runtime_flags);
btrfs_invalidate_inodes(btrfs_inode->root);
}
@@ -3683,13 +3794,9 @@ static int btrfs_destroy_marked_extents(struct btrfs_root *root,
int mark)
{
int ret;
- struct page *page;
- struct inode *btree_inode = root->fs_info->btree_inode;
struct extent_buffer *eb;
u64 start = 0;
u64 end;
- u64 offset;
- unsigned long index;
while (1) {
ret = find_first_extent_bit(dirty_pages, start, &start, &end,
@@ -3699,36 +3806,17 @@ static int btrfs_destroy_marked_extents(struct btrfs_root *root,
clear_extent_bits(dirty_pages, start, end, mark, GFP_NOFS);
while (start <= end) {
- index = start >> PAGE_CACHE_SHIFT;
- start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
- page = find_get_page(btree_inode->i_mapping, index);
- if (!page)
+ eb = btrfs_find_tree_block(root, start,
+ root->leafsize);
+ start += eb->len;
+ if (!eb)
continue;
- offset = page_offset(page);
-
- spin_lock(&dirty_pages->buffer_lock);
- eb = radix_tree_lookup(
- &(&BTRFS_I(page->mapping->host)->io_tree)->buffer,
- offset >> PAGE_CACHE_SHIFT);
- spin_unlock(&dirty_pages->buffer_lock);
- if (eb)
- ret = test_and_clear_bit(EXTENT_BUFFER_DIRTY,
- &eb->bflags);
- if (PageWriteback(page))
- end_page_writeback(page);
-
- lock_page(page);
- if (PageDirty(page)) {
- clear_page_dirty_for_io(page);
- spin_lock_irq(&page->mapping->tree_lock);
- radix_tree_tag_clear(&page->mapping->page_tree,
- page_index(page),
- PAGECACHE_TAG_DIRTY);
- spin_unlock_irq(&page->mapping->tree_lock);
- }
+ wait_on_extent_buffer_writeback(eb);
- unlock_page(page);
- page_cache_release(page);
+ if (test_and_clear_bit(EXTENT_BUFFER_DIRTY,
+ &eb->bflags))
+ clear_extent_buffer_dirty(eb);
+ free_extent_buffer_stale(eb);
}
}
@@ -3787,6 +3875,8 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
cur_trans->blocked = 1;
wake_up(&root->fs_info->transaction_blocked_wait);
+ btrfs_evict_pending_snapshots(cur_trans);
+
cur_trans->blocked = 0;
wake_up(&root->fs_info->transaction_wait);
@@ -3796,8 +3886,6 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
btrfs_destroy_delayed_inodes(root);
btrfs_assert_delayed_root_empty(root);
- btrfs_destroy_pending_snapshots(cur_trans);
-
btrfs_destroy_marked_extents(root, &cur_trans->dirty_pages,
EXTENT_DIRTY);
btrfs_destroy_pinned_extent(root,
@@ -3809,7 +3897,7 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
*/
}
-int btrfs_cleanup_transaction(struct btrfs_root *root)
+static int btrfs_cleanup_transaction(struct btrfs_root *root)
{
struct btrfs_transaction *t;
LIST_HEAD(list);
@@ -3823,19 +3911,13 @@ int btrfs_cleanup_transaction(struct btrfs_root *root)
while (!list_empty(&list)) {
t = list_entry(list.next, struct btrfs_transaction, list);
- if (!t)
- break;
- btrfs_destroy_ordered_operations(root);
+ btrfs_destroy_ordered_operations(t, root);
btrfs_destroy_ordered_extents(root);
btrfs_destroy_delayed_refs(t, root);
- btrfs_block_rsv_release(root,
- &root->fs_info->trans_block_rsv,
- t->dirty_pages.dirty_bytes);
-
/* FIXME: cleanup wait for commit */
t->in_commit = 1;
t->blocked = 1;
@@ -3843,6 +3925,8 @@ int btrfs_cleanup_transaction(struct btrfs_root *root)
if (waitqueue_active(&root->fs_info->transaction_blocked_wait))
wake_up(&root->fs_info->transaction_blocked_wait);
+ btrfs_evict_pending_snapshots(t);
+
t->blocked = 0;
smp_mb();
if (waitqueue_active(&root->fs_info->transaction_wait))
@@ -3856,8 +3940,6 @@ int btrfs_cleanup_transaction(struct btrfs_root *root)
btrfs_destroy_delayed_inodes(root);
btrfs_assert_delayed_root_empty(root);
- btrfs_destroy_pending_snapshots(t);
-
btrfs_destroy_delalloc_inodes(root);
spin_lock(&root->fs_info->trans_lock);
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 305c33efb0e..be69ce1b07a 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -25,6 +25,13 @@
#define BTRFS_SUPER_MIRROR_MAX 3
#define BTRFS_SUPER_MIRROR_SHIFT 12
+enum {
+ BTRFS_WQ_ENDIO_DATA = 0,
+ BTRFS_WQ_ENDIO_METADATA = 1,
+ BTRFS_WQ_ENDIO_FREE_SPACE = 2,
+ BTRFS_WQ_ENDIO_RAID56 = 3,
+};
+
static inline u64 btrfs_sb_offset(int mirror)
{
u64 start = 16 * 1024;
@@ -54,7 +61,6 @@ int write_ctree_super(struct btrfs_trans_handle *trans,
struct btrfs_root *root, int max_mirrors);
struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
int btrfs_commit_super(struct btrfs_root *root);
-void btrfs_error_commit_super(struct btrfs_root *root);
struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
u64 bytenr, u32 blocksize);
struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
@@ -70,7 +76,7 @@ int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
int atomic);
int btrfs_set_buffer_uptodate(struct extent_buffer *buf);
int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid);
-u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len);
+u32 btrfs_csum_data(char *data, u32 seed, size_t len);
void btrfs_csum_final(u32 crc, char *result);
int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
int metadata);
@@ -86,10 +92,8 @@ 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,
struct btrfs_root *root);
-int btrfs_cleanup_transaction(struct btrfs_root *root);
void btrfs_cleanup_one_transaction(struct btrfs_transaction *trans,
struct btrfs_root *root);
-void btrfs_abort_devices(struct btrfs_root *root);
struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
u64 objectid);
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 614f34a899c..81ee29eeb7c 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -22,10 +22,10 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
if (parent && (len < BTRFS_FID_SIZE_CONNECTABLE)) {
*max_len = BTRFS_FID_SIZE_CONNECTABLE;
- return 255;
+ return FILEID_INVALID;
} else if (len < BTRFS_FID_SIZE_NON_CONNECTABLE) {
*max_len = BTRFS_FID_SIZE_NON_CONNECTABLE;
- return 255;
+ return FILEID_INVALID;
}
len = BTRFS_FID_SIZE_NON_CONNECTABLE;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 5a3327b8f90..2305b5c5cf0 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -31,6 +31,7 @@
#include "print-tree.h"
#include "transaction.h"
#include "volumes.h"
+#include "raid56.h"
#include "locking.h"
#include "free-space-cache.h"
#include "math.h"
@@ -72,8 +73,7 @@ enum {
RESERVE_ALLOC_NO_ACCOUNT = 2,
};
-static int update_block_group(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+static int update_block_group(struct btrfs_root *root,
u64 bytenr, u64 num_bytes, int alloc);
static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -103,6 +103,10 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
int dump_block_groups);
static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
u64 num_bytes, int reserve);
+static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv,
+ u64 num_bytes);
+int btrfs_pin_extent(struct btrfs_root *root,
+ u64 bytenr, u64 num_bytes, int reserved);
static noinline int
block_group_cache_done(struct btrfs_block_group_cache *cache)
@@ -162,6 +166,10 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
rb_link_node(&block_group->cache_node, parent, p);
rb_insert_color(&block_group->cache_node,
&info->block_group_cache_tree);
+
+ if (info->first_logical_byte > block_group->key.objectid)
+ info->first_logical_byte = block_group->key.objectid;
+
spin_unlock(&info->block_group_cache_lock);
return 0;
@@ -203,8 +211,11 @@ block_group_cache_tree_search(struct btrfs_fs_info *info, u64 bytenr,
break;
}
}
- if (ret)
+ if (ret) {
btrfs_get_block_group(ret);
+ if (bytenr == 0 && info->first_logical_byte > ret->key.objectid)
+ info->first_logical_byte = ret->key.objectid;
+ }
spin_unlock(&info->block_group_cache_lock);
return ret;
@@ -248,7 +259,8 @@ static int exclude_super_stripes(struct btrfs_root *root,
cache->bytes_super += stripe_len;
ret = add_excluded_extent(root, cache->key.objectid,
stripe_len);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ return ret;
}
for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
@@ -256,13 +268,35 @@ static int exclude_super_stripes(struct btrfs_root *root,
ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
cache->key.objectid, bytenr,
0, &logical, &nr, &stripe_len);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ return ret;
while (nr--) {
- cache->bytes_super += stripe_len;
- ret = add_excluded_extent(root, logical[nr],
- stripe_len);
- BUG_ON(ret); /* -ENOMEM */
+ u64 start, len;
+
+ if (logical[nr] > cache->key.objectid +
+ cache->key.offset)
+ continue;
+
+ if (logical[nr] + stripe_len <= cache->key.objectid)
+ continue;
+
+ start = logical[nr];
+ if (start < cache->key.objectid) {
+ start = cache->key.objectid;
+ len = (logical[nr] + stripe_len) - start;
+ } else {
+ len = min_t(u64, stripe_len,
+ cache->key.objectid +
+ cache->key.offset - start);
+ }
+
+ cache->bytes_super += len;
+ ret = add_excluded_extent(root, start, len);
+ if (ret) {
+ kfree(logical);
+ return ret;
+ }
}
kfree(logical);
@@ -405,8 +439,7 @@ again:
if (ret)
break;
- if (need_resched() ||
- btrfs_next_leaf(extent_root, path)) {
+ if (need_resched()) {
caching_ctl->progress = last;
btrfs_release_path(path);
up_read(&fs_info->extent_commit_sem);
@@ -414,6 +447,12 @@ again:
cond_resched();
goto again;
}
+
+ ret = btrfs_next_leaf(extent_root, path);
+ if (ret < 0)
+ goto err;
+ if (ret)
+ break;
leaf = path->nodes[0];
nritems = btrfs_header_nritems(leaf);
continue;
@@ -428,11 +467,16 @@ again:
block_group->key.offset)
break;
- if (key.type == BTRFS_EXTENT_ITEM_KEY) {
+ if (key.type == BTRFS_EXTENT_ITEM_KEY ||
+ key.type == BTRFS_METADATA_ITEM_KEY) {
total_found += add_new_free_space(block_group,
fs_info, last,
key.objectid);
- last = key.objectid + key.offset;
+ if (key.type == BTRFS_METADATA_ITEM_KEY)
+ last = key.objectid +
+ fs_info->tree_root->leafsize;
+ else
+ last = key.objectid + key.offset;
if (total_found > (1024 * 1024 * 2)) {
total_found = 0;
@@ -468,8 +512,6 @@ out:
}
static int cache_block_group(struct btrfs_block_group_cache *cache,
- struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
int load_cache_only)
{
DEFINE_WAIT(wait);
@@ -527,12 +569,6 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
cache->cached = BTRFS_CACHE_FAST;
spin_unlock(&cache->lock);
- /*
- * We can't do the read from on-disk cache during a commit since we need
- * to have the normal tree locking. Also if we are currently trying to
- * allocate blocks for the tree root we can't do the fast caching since
- * we likely hold important locks.
- */
if (fs_info->mount_opt & BTRFS_MOUNT_SPACE_CACHE) {
ret = load_free_space_cache(fs_info, cache);
@@ -650,55 +686,6 @@ void btrfs_clear_space_info_full(struct btrfs_fs_info *info)
rcu_read_unlock();
}
-u64 btrfs_find_block_group(struct btrfs_root *root,
- u64 search_start, u64 search_hint, int owner)
-{
- struct btrfs_block_group_cache *cache;
- u64 used;
- u64 last = max(search_hint, search_start);
- u64 group_start = 0;
- int full_search = 0;
- int factor = 9;
- int wrapped = 0;
-again:
- while (1) {
- cache = btrfs_lookup_first_block_group(root->fs_info, last);
- if (!cache)
- break;
-
- spin_lock(&cache->lock);
- last = cache->key.objectid + cache->key.offset;
- used = btrfs_block_group_used(&cache->item);
-
- if ((full_search || !cache->ro) &&
- block_group_bits(cache, BTRFS_BLOCK_GROUP_METADATA)) {
- if (used + cache->pinned + cache->reserved <
- div_factor(cache->key.offset, factor)) {
- group_start = cache->key.objectid;
- spin_unlock(&cache->lock);
- btrfs_put_block_group(cache);
- goto found;
- }
- }
- spin_unlock(&cache->lock);
- btrfs_put_block_group(cache);
- cond_resched();
- }
- if (!wrapped) {
- last = search_start;
- wrapped = 1;
- goto again;
- }
- if (!full_search && factor < 10) {
- last = search_start;
- full_search = 1;
- factor = 10;
- goto again;
- }
-found:
- return group_start;
-}
-
/* simple helper to search for an existing extent at a given offset */
int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len)
{
@@ -712,15 +699,21 @@ int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len)
key.objectid = start;
key.offset = len;
- btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
+ key.type = BTRFS_EXTENT_ITEM_KEY;
ret = btrfs_search_slot(NULL, root->fs_info->extent_root, &key, path,
0, 0);
+ if (ret > 0) {
+ btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+ if (key.objectid == start &&
+ key.type == BTRFS_METADATA_ITEM_KEY)
+ ret = 0;
+ }
btrfs_free_path(path);
return ret;
}
/*
- * helper function to lookup reference count and flags of extent.
+ * helper function to lookup reference count and flags of a tree block.
*
* the head node for delayed ref is used to store the sum of all the
* reference count modifications queued up in the rbtree. the head
@@ -730,7 +723,7 @@ int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len)
*/
int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 bytenr,
- u64 num_bytes, u64 *refs, u64 *flags)
+ u64 offset, int metadata, u64 *refs, u64 *flags)
{
struct btrfs_delayed_ref_head *head;
struct btrfs_delayed_ref_root *delayed_refs;
@@ -743,13 +736,29 @@ int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
u64 extent_flags;
int ret;
+ /*
+ * If we don't have skinny metadata, don't bother doing anything
+ * different
+ */
+ if (metadata && !btrfs_fs_incompat(root->fs_info, SKINNY_METADATA)) {
+ offset = root->leafsize;
+ metadata = 0;
+ }
+
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
- key.objectid = bytenr;
- key.type = BTRFS_EXTENT_ITEM_KEY;
- key.offset = num_bytes;
+ if (metadata) {
+ key.objectid = bytenr;
+ key.type = BTRFS_METADATA_ITEM_KEY;
+ key.offset = offset;
+ } else {
+ key.objectid = bytenr;
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.offset = offset;
+ }
+
if (!trans) {
path->skip_locking = 1;
path->search_commit_root = 1;
@@ -760,6 +769,13 @@ again:
if (ret < 0)
goto out_free;
+ if (ret > 0 && metadata && key.type == BTRFS_METADATA_ITEM_KEY) {
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.offset = root->leafsize;
+ btrfs_release_path(path);
+ goto again;
+ }
+
if (ret == 0) {
leaf = path->nodes[0];
item_size = btrfs_item_size_nr(leaf, path->slots[0]);
@@ -995,7 +1011,7 @@ static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
return ret;
BUG_ON(ret); /* Corruption */
- btrfs_extend_item(trans, root, path, new_size);
+ btrfs_extend_item(root, path, new_size);
leaf = path->nodes[0];
item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
@@ -1447,6 +1463,8 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
int want;
int ret;
int err = 0;
+ bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
+ SKINNY_METADATA);
key.objectid = bytenr;
key.type = BTRFS_EXTENT_ITEM_KEY;
@@ -1458,16 +1476,54 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
path->keep_locks = 1;
} else
extra_size = -1;
+
+ /*
+ * Owner is our parent level, so we can just add one to get the level
+ * for the block we are interested in.
+ */
+ if (skinny_metadata && owner < BTRFS_FIRST_FREE_OBJECTID) {
+ key.type = BTRFS_METADATA_ITEM_KEY;
+ key.offset = owner;
+ }
+
+again:
ret = btrfs_search_slot(trans, root, &key, path, extra_size, 1);
if (ret < 0) {
err = ret;
goto out;
}
+
+ /*
+ * We may be a newly converted file system which still has the old fat
+ * extent entries for metadata, so try and see if we have one of those.
+ */
+ if (ret > 0 && skinny_metadata) {
+ skinny_metadata = false;
+ if (path->slots[0]) {
+ path->slots[0]--;
+ btrfs_item_key_to_cpu(path->nodes[0], &key,
+ path->slots[0]);
+ if (key.objectid == bytenr &&
+ key.type == BTRFS_EXTENT_ITEM_KEY &&
+ key.offset == num_bytes)
+ ret = 0;
+ }
+ if (ret) {
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.offset = num_bytes;
+ btrfs_release_path(path);
+ goto again;
+ }
+ }
+
if (ret && !insert) {
err = -ENOENT;
goto out;
+ } else if (ret) {
+ err = -EIO;
+ WARN_ON(1);
+ goto out;
}
- BUG_ON(ret); /* Corruption */
leaf = path->nodes[0];
item_size = btrfs_item_size_nr(leaf, path->slots[0]);
@@ -1495,11 +1551,9 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
ptr = (unsigned long)(ei + 1);
end = (unsigned long)ei + item_size;
- if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
+ if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK && !skinny_metadata) {
ptr += sizeof(struct btrfs_tree_block_info);
BUG_ON(ptr > end);
- } else {
- BUG_ON(!(flags & BTRFS_EXTENT_FLAG_DATA));
}
err = -ENOENT;
@@ -1581,8 +1635,7 @@ out:
* helper to add new inline back ref
*/
static noinline_for_stack
-void setup_inline_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+void setup_inline_extent_backref(struct btrfs_root *root,
struct btrfs_path *path,
struct btrfs_extent_inline_ref *iref,
u64 parent, u64 root_objectid,
@@ -1605,7 +1658,7 @@ void setup_inline_extent_backref(struct btrfs_trans_handle *trans,
type = extent_ref_type(parent, owner);
size = btrfs_extent_inline_ref_size(type);
- btrfs_extend_item(trans, root, path, size);
+ btrfs_extend_item(root, path, size);
ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
refs = btrfs_extent_refs(leaf, ei);
@@ -1674,8 +1727,7 @@ 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_trans_handle *trans,
- struct btrfs_root *root,
+void update_inline_extent_backref(struct btrfs_root *root,
struct btrfs_path *path,
struct btrfs_extent_inline_ref *iref,
int refs_to_mod,
@@ -1731,7 +1783,7 @@ void update_inline_extent_backref(struct btrfs_trans_handle *trans,
memmove_extent_buffer(leaf, ptr, ptr + size,
end - ptr - size);
item_size -= size;
- btrfs_truncate_item(trans, root, path, item_size, 1);
+ btrfs_truncate_item(root, path, item_size, 1);
}
btrfs_mark_buffer_dirty(leaf);
}
@@ -1753,10 +1805,10 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
root_objectid, owner, offset, 1);
if (ret == 0) {
BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID);
- update_inline_extent_backref(trans, root, path, iref,
+ update_inline_extent_backref(root, path, iref,
refs_to_add, extent_op);
} else if (ret == -ENOENT) {
- setup_inline_extent_backref(trans, root, path, iref, parent,
+ setup_inline_extent_backref(root, path, iref, parent,
root_objectid, owner, offset,
refs_to_add, extent_op);
ret = 0;
@@ -1793,7 +1845,7 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans,
BUG_ON(!is_data && refs_to_drop != 1);
if (iref) {
- update_inline_extent_backref(trans, root, path, iref,
+ update_inline_extent_backref(root, path, iref,
-refs_to_drop, NULL);
} else if (is_data) {
ret = remove_extent_data_ref(trans, root, path, refs_to_drop);
@@ -1852,6 +1904,8 @@ static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
*actual_bytes = discarded_bytes;
+ if (ret == -EOPNOTSUPP)
+ ret = 0;
return ret;
}
@@ -1962,10 +2016,8 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
ref_root = ref->root;
if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
- if (extent_op) {
- BUG_ON(extent_op->update_key);
+ if (extent_op)
flags |= extent_op->flags_to_set;
- }
ret = alloc_reserved_file_extent(trans, root,
parent, ref_root, flags,
ref->objectid, ref->offset,
@@ -2018,18 +2070,33 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
u32 item_size;
int ret;
int err = 0;
+ int metadata = (node->type == BTRFS_TREE_BLOCK_REF_KEY ||
+ node->type == BTRFS_SHARED_BLOCK_REF_KEY);
if (trans->aborted)
return 0;
+ if (metadata && !btrfs_fs_incompat(root->fs_info, SKINNY_METADATA))
+ metadata = 0;
+
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
key.objectid = node->bytenr;
- key.type = BTRFS_EXTENT_ITEM_KEY;
- key.offset = node->num_bytes;
+ if (metadata) {
+ struct btrfs_delayed_tree_ref *tree_ref;
+
+ tree_ref = btrfs_delayed_node_to_tree_ref(node);
+ key.type = BTRFS_METADATA_ITEM_KEY;
+ key.offset = tree_ref->level;
+ } else {
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.offset = node->num_bytes;
+ }
+
+again:
path->reada = 1;
path->leave_spinning = 1;
ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key,
@@ -2039,6 +2106,14 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
goto out;
}
if (ret > 0) {
+ if (metadata) {
+ btrfs_release_path(path);
+ metadata = 0;
+
+ key.offset = node->num_bytes;
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ goto again;
+ }
err = -EIO;
goto out;
}
@@ -2078,10 +2153,8 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
struct btrfs_key ins;
u64 parent = 0;
u64 ref_root = 0;
-
- ins.objectid = node->bytenr;
- ins.offset = node->num_bytes;
- ins.type = BTRFS_EXTENT_ITEM_KEY;
+ bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
+ SKINNY_METADATA);
ref = btrfs_delayed_node_to_tree_ref(node);
if (node->type == BTRFS_SHARED_BLOCK_REF_KEY)
@@ -2089,10 +2162,18 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
else
ref_root = ref->root;
+ ins.objectid = node->bytenr;
+ if (skinny_metadata) {
+ ins.offset = ref->level;
+ ins.type = BTRFS_METADATA_ITEM_KEY;
+ } else {
+ ins.offset = node->num_bytes;
+ ins.type = BTRFS_EXTENT_ITEM_KEY;
+ }
+
BUG_ON(node->ref_mod != 1);
if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
- BUG_ON(!extent_op || !extent_op->update_flags ||
- !extent_op->update_key);
+ BUG_ON(!extent_op || !extent_op->update_flags);
ret = alloc_reserved_tree_block(trans, root,
parent, ref_root,
extent_op->flags_to_set,
@@ -2143,7 +2224,6 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
node->num_bytes);
}
}
- mutex_unlock(&head->mutex);
return ret;
}
@@ -2258,7 +2338,7 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
* process of being added. Don't run this ref yet.
*/
list_del_init(&locked_ref->cluster);
- mutex_unlock(&locked_ref->mutex);
+ btrfs_delayed_ref_unlock(locked_ref);
locked_ref = NULL;
delayed_refs->num_heads_ready++;
spin_unlock(&delayed_refs->lock);
@@ -2285,7 +2365,7 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
ref = &locked_ref->node;
if (extent_op && must_insert_reserved) {
- kfree(extent_op);
+ btrfs_free_delayed_extent_op(extent_op);
extent_op = NULL;
}
@@ -2294,28 +2374,23 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
ret = run_delayed_extent_op(trans, root,
ref, extent_op);
- kfree(extent_op);
+ btrfs_free_delayed_extent_op(extent_op);
if (ret) {
- list_del_init(&locked_ref->cluster);
- mutex_unlock(&locked_ref->mutex);
-
- printk(KERN_DEBUG "btrfs: run_delayed_extent_op returned %d\n", ret);
+ btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret);
spin_lock(&delayed_refs->lock);
+ btrfs_delayed_ref_unlock(locked_ref);
return ret;
}
goto next;
}
-
- list_del_init(&locked_ref->cluster);
- locked_ref = NULL;
}
ref->in_tree = 0;
rb_erase(&ref->rb_node, &delayed_refs->root);
delayed_refs->num_entries--;
- if (locked_ref) {
+ if (!btrfs_delayed_ref_is_head(ref)) {
/*
* when we play the delayed ref, also correct the
* ref_mod on head
@@ -2337,20 +2412,28 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
ret = run_one_delayed_ref(trans, root, ref, extent_op,
must_insert_reserved);
- btrfs_put_delayed_ref(ref);
- kfree(extent_op);
- count++;
-
+ btrfs_free_delayed_extent_op(extent_op);
if (ret) {
- if (locked_ref) {
- list_del_init(&locked_ref->cluster);
- mutex_unlock(&locked_ref->mutex);
- }
- printk(KERN_DEBUG "btrfs: run_one_delayed_ref returned %d\n", ret);
+ btrfs_delayed_ref_unlock(locked_ref);
+ btrfs_put_delayed_ref(ref);
+ btrfs_debug(fs_info, "run_one_delayed_ref returned %d", ret);
spin_lock(&delayed_refs->lock);
return ret;
}
+ /*
+ * If this node is a head, that means all the refs in this head
+ * have been dealt with, and we will pick the next head to deal
+ * with, so we must unlock the head and drop it from the cluster
+ * list before we release it.
+ */
+ if (btrfs_delayed_ref_is_head(ref)) {
+ list_del_init(&locked_ref->cluster);
+ btrfs_delayed_ref_unlock(locked_ref);
+ locked_ref = NULL;
+ }
+ btrfs_put_delayed_ref(ref);
+ count++;
next:
cond_resched();
spin_lock(&delayed_refs->lock);
@@ -2410,9 +2493,11 @@ int btrfs_delayed_refs_qgroup_accounting(struct btrfs_trans_handle *trans,
if (list_empty(&trans->qgroup_ref_list) !=
!trans->delayed_ref_elem.seq) {
/* list without seq or seq without list */
- printk(KERN_ERR "btrfs: qgroup accounting update error, list is%s empty, seq is %llu\n",
+ btrfs_err(fs_info,
+ "qgroup accounting update error, list is%s empty, seq is %#x.%x",
list_empty(&trans->qgroup_ref_list) ? "" : " not",
- trans->delayed_ref_elem.seq);
+ (u32)(trans->delayed_ref_elem.seq >> 32),
+ (u32)trans->delayed_ref_elem.seq);
BUG();
}
@@ -2435,6 +2520,16 @@ int btrfs_delayed_refs_qgroup_accounting(struct btrfs_trans_handle *trans,
return ret;
}
+static int refs_newer(struct btrfs_delayed_ref_root *delayed_refs, int seq,
+ int count)
+{
+ int val = atomic_read(&delayed_refs->ref_seq);
+
+ if (val < seq || val >= seq + count)
+ return 1;
+ return 0;
+}
+
/*
* this starts processing the delayed reference count updates and
* extent insertions we have queued up so far. count can be
@@ -2469,6 +2564,44 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
delayed_refs = &trans->transaction->delayed_refs;
INIT_LIST_HEAD(&cluster);
+ if (count == 0) {
+ count = delayed_refs->num_entries * 2;
+ run_most = 1;
+ }
+
+ if (!run_all && !run_most) {
+ int old;
+ int seq = atomic_read(&delayed_refs->ref_seq);
+
+progress:
+ old = atomic_cmpxchg(&delayed_refs->procs_running_refs, 0, 1);
+ if (old) {
+ DEFINE_WAIT(__wait);
+ if (delayed_refs->num_entries < 16348)
+ return 0;
+
+ prepare_to_wait(&delayed_refs->wait, &__wait,
+ TASK_UNINTERRUPTIBLE);
+
+ old = atomic_cmpxchg(&delayed_refs->procs_running_refs, 0, 1);
+ if (old) {
+ schedule();
+ finish_wait(&delayed_refs->wait, &__wait);
+
+ if (!refs_newer(delayed_refs, seq, 256))
+ goto progress;
+ else
+ return 0;
+ } else {
+ finish_wait(&delayed_refs->wait, &__wait);
+ goto again;
+ }
+ }
+
+ } else {
+ atomic_inc(&delayed_refs->procs_running_refs);
+ }
+
again:
loops = 0;
spin_lock(&delayed_refs->lock);
@@ -2477,10 +2610,6 @@ again:
delayed_refs->run_delayed_start = find_middle(&delayed_refs->root);
#endif
- if (count == 0) {
- count = delayed_refs->num_entries * 2;
- run_most = 1;
- }
while (1) {
if (!(run_all || run_most) &&
delayed_refs->num_heads_ready < 64)
@@ -2500,11 +2629,15 @@ again:
ret = run_clustered_refs(trans, root, &cluster);
if (ret < 0) {
+ btrfs_release_ref_cluster(&cluster);
spin_unlock(&delayed_refs->lock);
btrfs_abort_transaction(trans, root, ret);
+ atomic_dec(&delayed_refs->procs_running_refs);
return ret;
}
+ atomic_add(ret, &delayed_refs->ref_seq);
+
count -= min_t(unsigned long, ret, count);
if (count == 0)
@@ -2573,6 +2706,11 @@ again:
goto again;
}
out:
+ atomic_dec(&delayed_refs->procs_running_refs);
+ smp_mb();
+ if (waitqueue_active(&delayed_refs->wait))
+ wake_up(&delayed_refs->wait);
+
spin_unlock(&delayed_refs->lock);
assert_qgroups_uptodate(trans);
return 0;
@@ -2586,7 +2724,7 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
struct btrfs_delayed_extent_op *extent_op;
int ret;
- extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
+ extent_op = btrfs_alloc_delayed_extent_op();
if (!extent_op)
return -ENOMEM;
@@ -2598,7 +2736,7 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
ret = btrfs_add_delayed_extent_op(root->fs_info, trans, bytenr,
num_bytes, extent_op);
if (ret)
- kfree(extent_op);
+ btrfs_free_delayed_extent_op(extent_op);
return ret;
}
@@ -3223,12 +3361,14 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
u64 extra_flags = chunk_to_extended(flags) &
BTRFS_EXTENDED_PROFILE_MASK;
+ write_seqlock(&fs_info->profiles_lock);
if (flags & BTRFS_BLOCK_GROUP_DATA)
fs_info->avail_data_alloc_bits |= extra_flags;
if (flags & BTRFS_BLOCK_GROUP_METADATA)
fs_info->avail_metadata_alloc_bits |= extra_flags;
if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
fs_info->avail_system_alloc_bits |= extra_flags;
+ write_sequnlock(&fs_info->profiles_lock);
}
/*
@@ -3266,7 +3406,7 @@ static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags)
* progress (either running or paused) picks the target profile (if it's
* already available), otherwise falls back to plain reducing.
*/
-u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
+static u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
{
/*
* we add in the count of missing devices because we want
@@ -3276,6 +3416,7 @@ u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
u64 num_devices = root->fs_info->fs_devices->rw_devices +
root->fs_info->fs_devices->missing_devices;
u64 target;
+ u64 tmp;
/*
* see if restripe for this chunk_type is in progress, if so
@@ -3292,40 +3433,48 @@ u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
}
spin_unlock(&root->fs_info->balance_lock);
+ /* First, mask out the RAID levels which aren't possible */
if (num_devices == 1)
- flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0);
+ flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID5);
+ if (num_devices < 3)
+ flags &= ~BTRFS_BLOCK_GROUP_RAID6;
if (num_devices < 4)
flags &= ~BTRFS_BLOCK_GROUP_RAID10;
- if ((flags & BTRFS_BLOCK_GROUP_DUP) &&
- (flags & (BTRFS_BLOCK_GROUP_RAID1 |
- BTRFS_BLOCK_GROUP_RAID10))) {
- flags &= ~BTRFS_BLOCK_GROUP_DUP;
- }
+ tmp = flags & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6 | BTRFS_BLOCK_GROUP_RAID10);
+ flags &= ~tmp;
- if ((flags & BTRFS_BLOCK_GROUP_RAID1) &&
- (flags & BTRFS_BLOCK_GROUP_RAID10)) {
- flags &= ~BTRFS_BLOCK_GROUP_RAID1;
- }
+ if (tmp & BTRFS_BLOCK_GROUP_RAID6)
+ tmp = BTRFS_BLOCK_GROUP_RAID6;
+ else if (tmp & BTRFS_BLOCK_GROUP_RAID5)
+ tmp = BTRFS_BLOCK_GROUP_RAID5;
+ else if (tmp & BTRFS_BLOCK_GROUP_RAID10)
+ tmp = BTRFS_BLOCK_GROUP_RAID10;
+ else if (tmp & BTRFS_BLOCK_GROUP_RAID1)
+ tmp = BTRFS_BLOCK_GROUP_RAID1;
+ else if (tmp & BTRFS_BLOCK_GROUP_RAID0)
+ tmp = BTRFS_BLOCK_GROUP_RAID0;
- if ((flags & BTRFS_BLOCK_GROUP_RAID0) &&
- ((flags & BTRFS_BLOCK_GROUP_RAID1) |
- (flags & BTRFS_BLOCK_GROUP_RAID10) |
- (flags & BTRFS_BLOCK_GROUP_DUP))) {
- flags &= ~BTRFS_BLOCK_GROUP_RAID0;
- }
-
- return extended_to_chunk(flags);
+ return extended_to_chunk(flags | tmp);
}
static u64 get_alloc_profile(struct btrfs_root *root, u64 flags)
{
- if (flags & BTRFS_BLOCK_GROUP_DATA)
- flags |= root->fs_info->avail_data_alloc_bits;
- else if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
- flags |= root->fs_info->avail_system_alloc_bits;
- else if (flags & BTRFS_BLOCK_GROUP_METADATA)
- flags |= root->fs_info->avail_metadata_alloc_bits;
+ unsigned seq;
+
+ do {
+ seq = read_seqbegin(&root->fs_info->profiles_lock);
+
+ if (flags & BTRFS_BLOCK_GROUP_DATA)
+ flags |= root->fs_info->avail_data_alloc_bits;
+ else if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
+ flags |= root->fs_info->avail_system_alloc_bits;
+ else if (flags & BTRFS_BLOCK_GROUP_METADATA)
+ flags |= root->fs_info->avail_metadata_alloc_bits;
+ } while (read_seqretry(&root->fs_info->profiles_lock, seq));
return btrfs_reduce_alloc_profile(root, flags);
}
@@ -3333,6 +3482,7 @@ static u64 get_alloc_profile(struct btrfs_root *root, u64 flags)
u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
{
u64 flags;
+ u64 ret;
if (data)
flags = BTRFS_BLOCK_GROUP_DATA;
@@ -3341,7 +3491,8 @@ u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
else
flags = BTRFS_BLOCK_GROUP_METADATA;
- return get_alloc_profile(root, flags);
+ ret = get_alloc_profile(root, flags);
+ return ret;
}
/*
@@ -3357,7 +3508,7 @@ int btrfs_check_data_free_space(struct inode *inode, u64 bytes)
int ret = 0, committed = 0, alloc_chunk = 1;
/* make sure bytes are sectorsize aligned */
- bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
+ bytes = ALIGN(bytes, root->sectorsize);
if (root == root->fs_info->tree_root ||
BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID) {
@@ -3452,7 +3603,7 @@ void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes)
struct btrfs_space_info *data_sinfo;
/* make sure bytes are sectorsize aligned */
- bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
+ bytes = ALIGN(bytes, root->sectorsize);
data_sinfo = root->fs_info->data_sinfo;
spin_lock(&data_sinfo->lock);
@@ -3475,6 +3626,11 @@ static void force_metadata_allocation(struct btrfs_fs_info *info)
rcu_read_unlock();
}
+static inline u64 calc_global_rsv_need_space(struct btrfs_block_rsv *global)
+{
+ return (global->size << 1);
+}
+
static int should_alloc_chunk(struct btrfs_root *root,
struct btrfs_space_info *sinfo, int force)
{
@@ -3492,7 +3648,7 @@ static int should_alloc_chunk(struct btrfs_root *root,
* global_rsv, it doesn't change except when the transaction commits.
*/
if (sinfo->flags & BTRFS_BLOCK_GROUP_METADATA)
- num_allocated += global_rsv->size;
+ num_allocated += calc_global_rsv_need_space(global_rsv);
/*
* in limited mode, we want to have some free space up to
@@ -3516,8 +3672,10 @@ static u64 get_system_chunk_thresh(struct btrfs_root *root, u64 type)
{
u64 num_dev;
- if (type & BTRFS_BLOCK_GROUP_RAID10 ||
- type & BTRFS_BLOCK_GROUP_RAID0)
+ if (type & (BTRFS_BLOCK_GROUP_RAID10 |
+ BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6))
num_dev = root->fs_info->fs_devices->rw_devices;
else if (type & BTRFS_BLOCK_GROUP_RAID1)
num_dev = 2;
@@ -3543,8 +3701,8 @@ static void check_system_chunk(struct btrfs_trans_handle *trans,
thresh = get_system_chunk_thresh(root, type);
if (left < thresh && btrfs_test_opt(root, ENOSPC_DEBUG)) {
- printk(KERN_INFO "left=%llu, need=%llu, flags=%llu\n",
- left, thresh, type);
+ btrfs_info(root->fs_info, "left=%llu, need=%llu, flags=%llu",
+ left, thresh, type);
dump_space_info(info, 0, 0);
}
@@ -3564,6 +3722,10 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
int wait_for_alloc = 0;
int ret = 0;
+ /* Don't re-enter if we're already allocating a chunk */
+ if (trans->allocating_chunk)
+ return -ENOSPC;
+
space_info = __find_space_info(extent_root->fs_info, flags);
if (!space_info) {
ret = update_space_info(extent_root->fs_info, flags,
@@ -3606,6 +3768,8 @@ again:
goto again;
}
+ trans->allocating_chunk = true;
+
/*
* If we have mixed data/metadata chunks we want to make sure we keep
* allocating mixed chunks instead of individual chunks.
@@ -3632,19 +3796,20 @@ again:
check_system_chunk(trans, extent_root, flags);
ret = btrfs_alloc_chunk(trans, extent_root, flags);
- if (ret < 0 && ret != -ENOSPC)
- goto out;
+ trans->allocating_chunk = false;
spin_lock(&space_info->lock);
+ if (ret < 0 && ret != -ENOSPC)
+ goto out;
if (ret)
space_info->full = 1;
else
ret = 1;
space_info->force_alloc = CHUNK_ALLOC_NO_FORCE;
+out:
space_info->chunk_alloc = 0;
spin_unlock(&space_info->lock);
-out:
mutex_unlock(&fs_info->chunk_mutex);
return ret;
}
@@ -3653,13 +3818,29 @@ static int can_overcommit(struct btrfs_root *root,
struct btrfs_space_info *space_info, u64 bytes,
enum btrfs_reserve_flush_enum flush)
{
+ struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv;
u64 profile = btrfs_get_alloc_profile(root, 0);
+ u64 space_size;
u64 avail;
u64 used;
+ u64 to_add;
used = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- space_info->bytes_may_use;
+ space_info->bytes_pinned + space_info->bytes_readonly;
+
+ /*
+ * We only want to allow over committing if we have lots of actual space
+ * free, but if we don't have enough space to handle the global reserve
+ * space then we could end up having a real enospc problem when trying
+ * to allocate a chunk or some other such important allocation.
+ */
+ spin_lock(&global_rsv->lock);
+ space_size = calc_global_rsv_need_space(global_rsv);
+ spin_unlock(&global_rsv->lock);
+ if (used + space_size >= space_info->total_bytes)
+ return 0;
+
+ used += space_info->bytes_may_use;
spin_lock(&root->fs_info->free_chunk_lock);
avail = root->fs_info->free_chunk_space;
@@ -3667,40 +3848,59 @@ static int can_overcommit(struct btrfs_root *root,
/*
* If we have dup, raid1 or raid10 then only half of the free
- * space is actually useable.
+ * space is actually useable. For raid56, the space info used
+ * doesn't include the parity drive, so we don't have to
+ * change the math
*/
if (profile & (BTRFS_BLOCK_GROUP_DUP |
BTRFS_BLOCK_GROUP_RAID1 |
BTRFS_BLOCK_GROUP_RAID10))
avail >>= 1;
+ to_add = space_info->total_bytes;
+
/*
* If we aren't flushing all things, let us overcommit up to
* 1/2th of the space. If we can flush, don't let us overcommit
* too much, let it overcommit up to 1/8 of the space.
*/
if (flush == BTRFS_RESERVE_FLUSH_ALL)
- avail >>= 3;
+ to_add >>= 3;
else
- avail >>= 1;
+ to_add >>= 1;
+
+ /*
+ * Limit the overcommit to the amount of free space we could possibly
+ * allocate for chunks.
+ */
+ to_add = min(avail, to_add);
- if (used + bytes < space_info->total_bytes + avail)
+ if (used + bytes < space_info->total_bytes + to_add)
return 1;
return 0;
}
-static int writeback_inodes_sb_nr_if_idle_safe(struct super_block *sb,
- unsigned long nr_pages,
- enum wb_reason reason)
+static void btrfs_writeback_inodes_sb_nr(struct btrfs_root *root,
+ unsigned long nr_pages)
{
- if (!writeback_in_progress(sb->s_bdi) &&
- down_read_trylock(&sb->s_umount)) {
- writeback_inodes_sb_nr(sb, nr_pages, reason);
- up_read(&sb->s_umount);
- return 1;
- }
+ struct super_block *sb = root->fs_info->sb;
+ int started;
- return 0;
+ /* If we can not start writeback, just sync all the delalloc file. */
+ started = try_to_writeback_inodes_sb_nr(sb, nr_pages,
+ WB_REASON_FS_FREE_SPACE);
+ if (!started) {
+ /*
+ * We needn't worry the filesystem going from r/w to r/o though
+ * we don't acquire ->s_umount mutex, because the filesystem
+ * should guarantee the delalloc inodes list be empty after
+ * the filesystem is readonly(all dirty pages are written to
+ * the disk).
+ */
+ btrfs_start_delalloc_inodes(root, 0);
+ if (!current->journal_info)
+ btrfs_wait_ordered_extents(root, 0);
+ }
}
/*
@@ -3724,7 +3924,8 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
space_info = block_rsv->space_info;
smp_mb();
- delalloc_bytes = root->fs_info->delalloc_bytes;
+ delalloc_bytes = percpu_counter_sum_positive(
+ &root->fs_info->delalloc_bytes);
if (delalloc_bytes == 0) {
if (trans)
return;
@@ -3735,10 +3936,7 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
while (delalloc_bytes && loops < 3) {
max_reclaim = min(delalloc_bytes, to_reclaim);
nr_pages = max_reclaim >> PAGE_CACHE_SHIFT;
- writeback_inodes_sb_nr_if_idle_safe(root->fs_info->sb,
- nr_pages,
- WB_REASON_FS_FREE_SPACE);
-
+ btrfs_writeback_inodes_sb_nr(root, nr_pages);
/*
* We need to wait for the async pages to actually start before
* we do anything.
@@ -3766,7 +3964,8 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
break;
}
smp_mb();
- delalloc_bytes = root->fs_info->delalloc_bytes;
+ delalloc_bytes = percpu_counter_sum_positive(
+ &root->fs_info->delalloc_bytes);
}
}
@@ -4030,6 +4229,15 @@ again:
goto again;
out:
+ if (ret == -ENOSPC &&
+ unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
+ struct btrfs_block_rsv *global_rsv =
+ &root->fs_info->global_block_rsv;
+
+ if (block_rsv != global_rsv &&
+ !block_rsv_use_bytes(global_rsv, orig_bytes))
+ ret = 0;
+ }
if (flushing) {
spin_lock(&space_info->lock);
space_info->flush = 0;
@@ -4308,7 +4516,7 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
spin_lock(&sinfo->lock);
spin_lock(&block_rsv->lock);
- block_rsv->size = num_bytes;
+ block_rsv->size = min_t(u64, num_bytes, 512 * 1024 * 1024);
num_bytes = sinfo->bytes_used + sinfo->bytes_pinned +
sinfo->bytes_reserved + sinfo->bytes_readonly +
@@ -4416,19 +4624,60 @@ void btrfs_orphan_release_metadata(struct inode *inode)
btrfs_block_rsv_release(root, root->orphan_block_rsv, num_bytes);
}
-int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending)
+/*
+ * btrfs_subvolume_reserve_metadata() - reserve space for subvolume operation
+ * root: the root of the parent directory
+ * rsv: block reservation
+ * items: the number of items that we need do reservation
+ * qgroup_reserved: used to return the reserved size in qgroup
+ *
+ * This function is used to reserve the space for snapshot/subvolume
+ * creation and deletion. Those operations are different with the
+ * common file/directory operations, they change two fs/file trees
+ * and root tree, the number of items that the qgroup reserves is
+ * different with the free space reservation. So we can not use
+ * the space reseravtion mechanism in start_transaction().
+ */
+int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
+ struct btrfs_block_rsv *rsv,
+ int items,
+ u64 *qgroup_reserved)
+{
+ u64 num_bytes;
+ int ret;
+
+ if (root->fs_info->quota_enabled) {
+ /* One for parent inode, two for dir entries */
+ num_bytes = 3 * root->leafsize;
+ ret = btrfs_qgroup_reserve(root, num_bytes);
+ if (ret)
+ return ret;
+ } else {
+ num_bytes = 0;
+ }
+
+ *qgroup_reserved = num_bytes;
+
+ num_bytes = btrfs_calc_trans_metadata_size(root, items);
+ rsv->space_info = __find_space_info(root->fs_info,
+ BTRFS_BLOCK_GROUP_METADATA);
+ ret = btrfs_block_rsv_add(root, rsv, num_bytes,
+ BTRFS_RESERVE_FLUSH_ALL);
+ if (ret) {
+ if (*qgroup_reserved)
+ btrfs_qgroup_free(root, *qgroup_reserved);
+ }
+
+ return ret;
+}
+
+void btrfs_subvolume_release_metadata(struct btrfs_root *root,
+ struct btrfs_block_rsv *rsv,
+ u64 qgroup_reserved)
{
- struct btrfs_root *root = pending->root;
- struct btrfs_block_rsv *src_rsv = get_block_rsv(trans, root);
- struct btrfs_block_rsv *dst_rsv = &pending->block_rsv;
- /*
- * two for root back/forward refs, two for directory entries,
- * one for root of the snapshot and one for parent inode.
- */
- u64 num_bytes = btrfs_calc_trans_metadata_size(root, 6);
- dst_rsv->space_info = src_rsv->space_info;
- return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
+ btrfs_block_rsv_release(root, rsv, (u64)-1);
+ if (qgroup_reserved)
+ btrfs_qgroup_free(root, qgroup_reserved);
}
/**
@@ -4536,6 +4785,8 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
int ret = 0;
bool delalloc_lock = true;
+ u64 to_free = 0;
+ unsigned dropped;
/* If we are a free space inode we need to not flush since we will be in
* the middle of a transaction commit. We also don't need the delalloc
@@ -4579,54 +4830,19 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
csum_bytes = BTRFS_I(inode)->csum_bytes;
spin_unlock(&BTRFS_I(inode)->lock);
- if (root->fs_info->quota_enabled)
+ if (root->fs_info->quota_enabled) {
ret = btrfs_qgroup_reserve(root, num_bytes +
nr_extents * root->leafsize);
+ if (ret)
+ goto out_fail;
+ }
- /*
- * ret != 0 here means the qgroup reservation failed, we go straight to
- * the shared error handling then.
- */
- if (ret == 0)
- ret = reserve_metadata_bytes(root, block_rsv,
- to_reserve, flush);
-
- if (ret) {
- u64 to_free = 0;
- unsigned dropped;
-
- spin_lock(&BTRFS_I(inode)->lock);
- dropped = drop_outstanding_extent(inode);
- /*
- * 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.
- * Otherwise we have to do the normal free thing to account for
- * the case that the free side didn't free up its reserve
- * because of this outstanding reservation.
- */
- if (BTRFS_I(inode)->csum_bytes == csum_bytes)
- calc_csum_metadata_size(inode, num_bytes, 0);
- else
- to_free = calc_csum_metadata_size(inode, num_bytes, 0);
- spin_unlock(&BTRFS_I(inode)->lock);
- if (dropped)
- to_free += btrfs_calc_trans_metadata_size(root, dropped);
-
- if (to_free) {
- btrfs_block_rsv_release(root, block_rsv, to_free);
- trace_btrfs_space_reservation(root->fs_info,
- "delalloc",
- btrfs_ino(inode),
- to_free, 0);
- }
- if (root->fs_info->quota_enabled) {
+ ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush);
+ if (unlikely(ret)) {
+ if (root->fs_info->quota_enabled)
btrfs_qgroup_free(root, num_bytes +
nr_extents * root->leafsize);
- }
- if (delalloc_lock)
- mutex_unlock(&BTRFS_I(inode)->delalloc_mutex);
- return ret;
+ goto out_fail;
}
spin_lock(&BTRFS_I(inode)->lock);
@@ -4647,6 +4863,69 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
block_rsv_add_bytes(block_rsv, to_reserve, 1);
return 0;
+
+out_fail:
+ spin_lock(&BTRFS_I(inode)->lock);
+ dropped = drop_outstanding_extent(inode);
+ /*
+ * 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) {
+ calc_csum_metadata_size(inode, num_bytes, 0);
+ } else {
+ u64 orig_csum_bytes = BTRFS_I(inode)->csum_bytes;
+ u64 bytes;
+
+ /*
+ * This is tricky, but first we need to figure out how much we
+ * free'd from any free-ers that occured during this
+ * reservation, so we reset ->csum_bytes to the csum_bytes
+ * before we dropped our lock, and then call the free for the
+ * 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;
+ to_free = calc_csum_metadata_size(inode, bytes, 0);
+
+
+ /*
+ * Now we need to see how much we would have freed had we not
+ * been making this reservation and our ->csum_bytes were not
+ * artificially inflated.
+ */
+ BTRFS_I(inode)->csum_bytes = csum_bytes - num_bytes;
+ bytes = csum_bytes - orig_csum_bytes;
+ bytes = calc_csum_metadata_size(inode, bytes, 0);
+
+ /*
+ * Now reset ->csum_bytes to what it should be. If bytes is
+ * more than to_free then we would have free'd more space had we
+ * not had an artificially high ->csum_bytes, so we need to free
+ * the remainder. If bytes is the same or less then we don't
+ * need to do anything, the other free-ers did the correct
+ * thing.
+ */
+ BTRFS_I(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);
+ if (dropped)
+ to_free += btrfs_calc_trans_metadata_size(root, dropped);
+
+ if (to_free) {
+ btrfs_block_rsv_release(root, block_rsv, to_free);
+ trace_btrfs_space_reservation(root->fs_info, "delalloc",
+ btrfs_ino(inode), to_free, 0);
+ }
+ if (delalloc_lock)
+ mutex_unlock(&BTRFS_I(inode)->delalloc_mutex);
+ return ret;
}
/**
@@ -4668,7 +4947,8 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
spin_lock(&BTRFS_I(inode)->lock);
dropped = drop_outstanding_extent(inode);
- to_free = calc_csum_metadata_size(inode, num_bytes, 0);
+ if (num_bytes)
+ to_free = calc_csum_metadata_size(inode, num_bytes, 0);
spin_unlock(&BTRFS_I(inode)->lock);
if (dropped > 0)
to_free += btrfs_calc_trans_metadata_size(root, dropped);
@@ -4735,8 +5015,7 @@ void btrfs_delalloc_release_space(struct inode *inode, u64 num_bytes)
btrfs_free_reserved_data_space(inode, num_bytes);
}
-static int update_block_group(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+static int update_block_group(struct btrfs_root *root,
u64 bytenr, u64 num_bytes, int alloc)
{
struct btrfs_block_group_cache *cache = NULL;
@@ -4773,7 +5052,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
* space back to the block group, otherwise we will leak space.
*/
if (!alloc && cache->cached == BTRFS_CACHE_NO)
- cache_block_group(cache, trans, NULL, 1);
+ cache_block_group(cache, 1);
byte_in_group = bytenr - cache->key.objectid;
WARN_ON(byte_in_group > cache->key.offset);
@@ -4823,6 +5102,13 @@ static u64 first_logical_byte(struct btrfs_root *root, u64 search_start)
struct btrfs_block_group_cache *cache;
u64 bytenr;
+ spin_lock(&root->fs_info->block_group_cache_lock);
+ bytenr = root->fs_info->first_logical_byte;
+ spin_unlock(&root->fs_info->block_group_cache_lock);
+
+ if (bytenr < (u64)-1)
+ return bytenr;
+
cache = btrfs_lookup_first_block_group(root->fs_info, search_start);
if (!cache)
return 0;
@@ -4873,14 +5159,15 @@ int btrfs_pin_extent(struct btrfs_root *root,
/*
* this function must be called within transaction
*/
-int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+int btrfs_pin_extent_for_log_replay(struct btrfs_root *root,
u64 bytenr, u64 num_bytes)
{
struct btrfs_block_group_cache *cache;
+ int ret;
cache = btrfs_lookup_block_group(root->fs_info, bytenr);
- BUG_ON(!cache); /* Logic error */
+ if (!cache)
+ return -EINVAL;
/*
* pull in the free space cache (if any) so that our pin
@@ -4888,14 +5175,14 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
* to one because the slow code to read in the free extents does check
* the pinned extents.
*/
- cache_block_group(cache, trans, root, 1);
+ cache_block_group(cache, 1);
pin_down_extent(root, cache, bytenr, num_bytes, 0);
/* remove us from the free space cache (if we're there at all) */
- btrfs_remove_free_space(cache, bytenr, num_bytes);
+ ret = btrfs_remove_free_space(cache, bytenr, num_bytes);
btrfs_put_block_group(cache);
- return 0;
+ return ret;
}
/**
@@ -5100,6 +5387,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
int num_to_del = 1;
u32 item_size;
u64 refs;
+ bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
+ SKINNY_METADATA);
path = btrfs_alloc_path();
if (!path)
@@ -5111,6 +5400,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
is_data = owner_objectid >= BTRFS_FIRST_FREE_OBJECTID;
BUG_ON(!is_data && refs_to_drop != 1);
+ if (is_data)
+ skinny_metadata = 0;
+
ret = lookup_extent_backref(trans, extent_root, path, &iref,
bytenr, num_bytes, parent,
root_objectid, owner_objectid,
@@ -5127,6 +5419,11 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
found_extent = 1;
break;
}
+ if (key.type == BTRFS_METADATA_ITEM_KEY &&
+ key.offset == owner_objectid) {
+ found_extent = 1;
+ break;
+ }
if (path->slots[0] - extent_slot > 5)
break;
extent_slot--;
@@ -5152,12 +5449,39 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
key.type = BTRFS_EXTENT_ITEM_KEY;
key.offset = num_bytes;
+ if (!is_data && skinny_metadata) {
+ key.type = BTRFS_METADATA_ITEM_KEY;
+ key.offset = owner_objectid;
+ }
+
ret = btrfs_search_slot(trans, extent_root,
&key, path, -1, 1);
+ if (ret > 0 && skinny_metadata && path->slots[0]) {
+ /*
+ * Couldn't find our skinny metadata item,
+ * see if we have ye olde extent item.
+ */
+ path->slots[0]--;
+ btrfs_item_key_to_cpu(path->nodes[0], &key,
+ path->slots[0]);
+ if (key.objectid == bytenr &&
+ key.type == BTRFS_EXTENT_ITEM_KEY &&
+ key.offset == num_bytes)
+ ret = 0;
+ }
+
+ if (ret > 0 && skinny_metadata) {
+ skinny_metadata = false;
+ key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.offset = num_bytes;
+ btrfs_release_path(path);
+ ret = btrfs_search_slot(trans, extent_root,
+ &key, path, -1, 1);
+ }
+
if (ret) {
- printk(KERN_ERR "umm, got %d back from search"
- ", was looking for %llu\n", ret,
- (unsigned long long)bytenr);
+ btrfs_err(info, "umm, got %d back from search, was looking for %llu",
+ ret, (unsigned long long)bytenr);
if (ret > 0)
btrfs_print_leaf(extent_root,
path->nodes[0]);
@@ -5171,13 +5495,13 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
} else if (ret == -ENOENT) {
btrfs_print_leaf(extent_root, path->nodes[0]);
WARN_ON(1);
- printk(KERN_ERR "btrfs unable to find ref byte nr %llu "
- "parent %llu root %llu owner %llu offset %llu\n",
- (unsigned long long)bytenr,
- (unsigned long long)parent,
- (unsigned long long)root_objectid,
- (unsigned long long)owner_objectid,
- (unsigned long long)owner_offset);
+ btrfs_err(info,
+ "unable to find ref byte nr %llu parent %llu root %llu owner %llu offset %llu",
+ (unsigned long long)bytenr,
+ (unsigned long long)parent,
+ (unsigned long long)root_objectid,
+ (unsigned long long)owner_objectid,
+ (unsigned long long)owner_offset);
} else {
btrfs_abort_transaction(trans, extent_root, ret);
goto out;
@@ -5205,9 +5529,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
ret = btrfs_search_slot(trans, extent_root, &key, path,
-1, 1);
if (ret) {
- printk(KERN_ERR "umm, got %d back from search"
- ", was looking for %llu\n", ret,
- (unsigned long long)bytenr);
+ btrfs_err(info, "umm, got %d back from search, was looking for %llu",
+ ret, (unsigned long long)bytenr);
btrfs_print_leaf(extent_root, path->nodes[0]);
}
if (ret < 0) {
@@ -5223,7 +5546,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
BUG_ON(item_size < sizeof(*ei));
ei = btrfs_item_ptr(leaf, extent_slot,
struct btrfs_extent_item);
- if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID) {
+ if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID &&
+ key.type == BTRFS_EXTENT_ITEM_KEY) {
struct btrfs_tree_block_info *bi;
BUG_ON(item_size < sizeof(*ei) + sizeof(*bi));
bi = (struct btrfs_tree_block_info *)(ei + 1);
@@ -5231,7 +5555,13 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
}
refs = btrfs_extent_refs(leaf, ei);
- BUG_ON(refs < refs_to_drop);
+ if (refs < refs_to_drop) {
+ btrfs_err(info, "trying to drop %d refs but we only have %Lu "
+ "for bytenr %Lu\n", refs_to_drop, refs, bytenr);
+ ret = -EINVAL;
+ btrfs_abort_transaction(trans, extent_root, ret);
+ goto out;
+ }
refs -= refs_to_drop;
if (refs > 0) {
@@ -5285,7 +5615,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
}
}
- ret = update_block_group(trans, root, bytenr, num_bytes, 0);
+ ret = update_block_group(root, bytenr, num_bytes, 0);
if (ret) {
btrfs_abort_transaction(trans, extent_root, ret);
goto out;
@@ -5330,7 +5660,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans,
if (head->extent_op) {
if (!head->must_insert_reserved)
goto out;
- kfree(head->extent_op);
+ btrfs_free_delayed_extent_op(head->extent_op);
head->extent_op = NULL;
}
@@ -5453,10 +5783,11 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_root *root,
return ret;
}
-static u64 stripe_align(struct btrfs_root *root, u64 val)
+static u64 stripe_align(struct btrfs_root *root,
+ struct btrfs_block_group_cache *cache,
+ u64 val, u64 num_bytes)
{
- u64 mask = ((u64)root->stripesize - 1);
- u64 ret = (val + mask) & ~mask;
+ u64 ret = ALIGN(val, root->stripesize);
return ret;
}
@@ -5476,7 +5807,6 @@ wait_block_group_cache_progress(struct btrfs_block_group_cache *cache,
u64 num_bytes)
{
struct btrfs_caching_control *caching_ctl;
- DEFINE_WAIT(wait);
caching_ctl = get_caching_control(cache);
if (!caching_ctl)
@@ -5493,7 +5823,6 @@ static noinline int
wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
{
struct btrfs_caching_control *caching_ctl;
- DEFINE_WAIT(wait);
caching_ctl = get_caching_control(cache);
if (!caching_ctl)
@@ -5507,20 +5836,20 @@ wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
int __get_raid_index(u64 flags)
{
- int index;
-
if (flags & BTRFS_BLOCK_GROUP_RAID10)
- index = 0;
+ return BTRFS_RAID_RAID10;
else if (flags & BTRFS_BLOCK_GROUP_RAID1)
- index = 1;
+ return BTRFS_RAID_RAID1;
else if (flags & BTRFS_BLOCK_GROUP_DUP)
- index = 2;
+ return BTRFS_RAID_DUP;
else if (flags & BTRFS_BLOCK_GROUP_RAID0)
- index = 3;
- else
- index = 4;
+ return BTRFS_RAID_RAID0;
+ else if (flags & BTRFS_BLOCK_GROUP_RAID5)
+ return BTRFS_RAID_RAID5;
+ else if (flags & BTRFS_BLOCK_GROUP_RAID6)
+ return BTRFS_RAID_RAID6;
- return index;
+ return BTRFS_RAID_SINGLE; /* BTRFS_BLOCK_GROUP_SINGLE */
}
static int get_block_group_index(struct btrfs_block_group_cache *cache)
@@ -5547,7 +5876,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *orig_root,
u64 num_bytes, u64 empty_size,
u64 hint_byte, struct btrfs_key *ins,
- u64 data)
+ u64 flags)
{
int ret = 0;
struct btrfs_root *root = orig_root->fs_info->extent_root;
@@ -5558,8 +5887,8 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
int empty_cluster = 2 * 1024 * 1024;
struct btrfs_space_info *space_info;
int loop = 0;
- int index = __get_raid_index(data);
- int alloc_type = (data & BTRFS_BLOCK_GROUP_DATA) ?
+ int index = __get_raid_index(flags);
+ int alloc_type = (flags & BTRFS_BLOCK_GROUP_DATA) ?
RESERVE_ALLOC_NO_ACCOUNT : RESERVE_ALLOC;
bool found_uncached_bg = false;
bool failed_cluster_refill = false;
@@ -5572,11 +5901,11 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
ins->objectid = 0;
ins->offset = 0;
- trace_find_free_extent(orig_root, num_bytes, empty_size, data);
+ trace_find_free_extent(orig_root, num_bytes, empty_size, flags);
- space_info = __find_space_info(root->fs_info, data);
+ space_info = __find_space_info(root->fs_info, flags);
if (!space_info) {
- printk(KERN_ERR "No space info for %llu\n", data);
+ btrfs_err(root->fs_info, "No space info for %llu", flags);
return -ENOSPC;
}
@@ -5587,13 +5916,13 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
if (btrfs_mixed_space_info(space_info))
use_cluster = false;
- if (data & BTRFS_BLOCK_GROUP_METADATA && use_cluster) {
+ if (flags & BTRFS_BLOCK_GROUP_METADATA && use_cluster) {
last_ptr = &root->fs_info->meta_alloc_cluster;
if (!btrfs_test_opt(root, SSD))
empty_cluster = 64 * 1024;
}
- if ((data & BTRFS_BLOCK_GROUP_DATA) && use_cluster &&
+ if ((flags & BTRFS_BLOCK_GROUP_DATA) && use_cluster &&
btrfs_test_opt(root, SSD)) {
last_ptr = &root->fs_info->data_alloc_cluster;
}
@@ -5622,7 +5951,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
* However if we are re-searching with an ideal block group
* picked out then we don't care that the block group is cached.
*/
- if (block_group && block_group_bits(block_group, data) &&
+ if (block_group && block_group_bits(block_group, flags) &&
block_group->cached != BTRFS_CACHE_NO) {
down_read(&space_info->groups_sem);
if (list_empty(&block_group->list) ||
@@ -5660,9 +5989,11 @@ search:
* raid types, but we want to make sure we only allocate
* for the proper type.
*/
- if (!block_group_bits(block_group, data)) {
+ if (!block_group_bits(block_group, flags)) {
u64 extra = BTRFS_BLOCK_GROUP_DUP |
BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6 |
BTRFS_BLOCK_GROUP_RAID10;
/*
@@ -5670,7 +6001,7 @@ search:
* doesn't provide them, bail. This does allow us to
* fill raid0 from raid1.
*/
- if ((data & extra) && !(block_group->flags & extra))
+ if ((flags & extra) && !(block_group->flags & extra))
goto loop;
}
@@ -5678,8 +6009,7 @@ have_block_group:
cached = block_group_cache_done(block_group);
if (unlikely(!cached)) {
found_uncached_bg = true;
- ret = cache_block_group(block_group, trans,
- orig_root, 0);
+ ret = cache_block_group(block_group, 0);
BUG_ON(ret < 0);
ret = 0;
}
@@ -5692,6 +6022,7 @@ have_block_group:
* lets look there
*/
if (last_ptr) {
+ unsigned long aligned_cluster;
/*
* the refill lock keeps out other
* people trying to start a new cluster
@@ -5701,7 +6032,7 @@ have_block_group:
if (used_block_group != block_group &&
(!used_block_group ||
used_block_group->ro ||
- !block_group_bits(used_block_group, data))) {
+ !block_group_bits(used_block_group, flags))) {
used_block_group = block_group;
goto refill_cluster;
}
@@ -5758,11 +6089,15 @@ refill_cluster:
goto unclustered_alloc;
}
+ aligned_cluster = max_t(unsigned long,
+ empty_cluster + empty_size,
+ block_group->full_stripe_len);
+
/* allocate a cluster in this block group */
ret = btrfs_find_space_cluster(trans, root,
block_group, last_ptr,
search_start, num_bytes,
- empty_cluster + empty_size);
+ aligned_cluster);
if (ret == 0) {
/*
* now pull our allocation out of this
@@ -5833,7 +6168,8 @@ unclustered_alloc:
goto loop;
}
checks:
- search_start = stripe_align(root, offset);
+ search_start = stripe_align(root, used_block_group,
+ offset, num_bytes);
/* move on to the next group */
if (search_start + num_bytes >
@@ -5892,7 +6228,7 @@ loop:
index = 0;
loop++;
if (loop == LOOP_ALLOC_CHUNK) {
- ret = do_chunk_alloc(trans, root, data,
+ ret = do_chunk_alloc(trans, root, flags,
CHUNK_ALLOC_FORCE);
/*
* Do not bail out on ENOSPC since we
@@ -5970,21 +6306,22 @@ int btrfs_reserve_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 num_bytes, u64 min_alloc_size,
u64 empty_size, u64 hint_byte,
- struct btrfs_key *ins, u64 data)
+ struct btrfs_key *ins, int is_data)
{
bool final_tried = false;
+ u64 flags;
int ret;
- data = btrfs_get_alloc_profile(root, data);
+ flags = btrfs_get_alloc_profile(root, is_data);
again:
WARN_ON(num_bytes < root->sectorsize);
ret = find_free_extent(trans, root, num_bytes, empty_size,
- hint_byte, ins, data);
+ hint_byte, ins, flags);
if (ret == -ENOSPC) {
if (!final_tried) {
num_bytes = num_bytes >> 1;
- num_bytes = num_bytes & ~(root->sectorsize - 1);
+ num_bytes = round_down(num_bytes, root->sectorsize);
num_bytes = max(num_bytes, min_alloc_size);
if (num_bytes == min_alloc_size)
final_tried = true;
@@ -5992,10 +6329,10 @@ again:
} else if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
struct btrfs_space_info *sinfo;
- sinfo = __find_space_info(root->fs_info, data);
- printk(KERN_ERR "btrfs allocation failed flags %llu, "
- "wanted %llu\n", (unsigned long long)data,
- (unsigned long long)num_bytes);
+ sinfo = __find_space_info(root->fs_info, flags);
+ btrfs_err(root->fs_info, "allocation failed flags %llu, wanted %llu",
+ (unsigned long long)flags,
+ (unsigned long long)num_bytes);
if (sinfo)
dump_space_info(sinfo, num_bytes, 1);
}
@@ -6014,8 +6351,8 @@ static int __btrfs_free_reserved_extent(struct btrfs_root *root,
cache = btrfs_lookup_block_group(root->fs_info, start);
if (!cache) {
- printk(KERN_ERR "Unable to find block group for %llu\n",
- (unsigned long long)start);
+ btrfs_err(root->fs_info, "Unable to find block group for %llu",
+ (unsigned long long)start);
return -ENOSPC;
}
@@ -6108,11 +6445,11 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_free_path(path);
- ret = update_block_group(trans, root, ins->objectid, ins->offset, 1);
+ ret = update_block_group(root, ins->objectid, ins->offset, 1);
if (ret) { /* -ENOENT, logic error */
- printk(KERN_ERR "btrfs update block group failed for %llu "
- "%llu\n", (unsigned long long)ins->objectid,
- (unsigned long long)ins->offset);
+ btrfs_err(fs_info, "update block group failed for %llu %llu",
+ (unsigned long long)ins->objectid,
+ (unsigned long long)ins->offset);
BUG();
}
return ret;
@@ -6131,7 +6468,12 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
struct btrfs_extent_inline_ref *iref;
struct btrfs_path *path;
struct extent_buffer *leaf;
- u32 size = sizeof(*extent_item) + sizeof(*block_info) + sizeof(*iref);
+ u32 size = sizeof(*extent_item) + sizeof(*iref);
+ bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
+ SKINNY_METADATA);
+
+ if (!skinny_metadata)
+ size += sizeof(*block_info);
path = btrfs_alloc_path();
if (!path)
@@ -6152,12 +6494,16 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
btrfs_set_extent_generation(leaf, extent_item, trans->transid);
btrfs_set_extent_flags(leaf, extent_item,
flags | BTRFS_EXTENT_FLAG_TREE_BLOCK);
- block_info = (struct btrfs_tree_block_info *)(extent_item + 1);
- btrfs_set_tree_block_key(leaf, block_info, key);
- btrfs_set_tree_block_level(leaf, block_info, level);
+ if (skinny_metadata) {
+ iref = (struct btrfs_extent_inline_ref *)(extent_item + 1);
+ } else {
+ block_info = (struct btrfs_tree_block_info *)(extent_item + 1);
+ btrfs_set_tree_block_key(leaf, block_info, key);
+ btrfs_set_tree_block_level(leaf, block_info, level);
+ iref = (struct btrfs_extent_inline_ref *)(block_info + 1);
+ }
- iref = (struct btrfs_extent_inline_ref *)(block_info + 1);
if (parent > 0) {
BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
btrfs_set_extent_inline_ref_type(leaf, iref,
@@ -6172,11 +6518,11 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
btrfs_mark_buffer_dirty(leaf);
btrfs_free_path(path);
- ret = update_block_group(trans, root, ins->objectid, ins->offset, 1);
+ ret = update_block_group(root, ins->objectid, root->leafsize, 1);
if (ret) { /* -ENOENT, logic error */
- printk(KERN_ERR "btrfs update block group failed for %llu "
- "%llu\n", (unsigned long long)ins->objectid,
- (unsigned long long)ins->offset);
+ btrfs_err(fs_info, "update block group failed for %llu %llu",
+ (unsigned long long)ins->objectid,
+ (unsigned long long)ins->offset);
BUG();
}
return ret;
@@ -6215,53 +6561,54 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
u64 num_bytes = ins->offset;
block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid);
- cache_block_group(block_group, trans, NULL, 0);
+ cache_block_group(block_group, 0);
caching_ctl = get_caching_control(block_group);
if (!caching_ctl) {
BUG_ON(!block_group_cache_done(block_group));
ret = btrfs_remove_free_space(block_group, start, num_bytes);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ goto out;
} else {
mutex_lock(&caching_ctl->mutex);
if (start >= caching_ctl->progress) {
ret = add_excluded_extent(root, start, num_bytes);
- BUG_ON(ret); /* -ENOMEM */
} else if (start + num_bytes <= caching_ctl->progress) {
ret = btrfs_remove_free_space(block_group,
start, num_bytes);
- BUG_ON(ret); /* -ENOMEM */
} else {
num_bytes = caching_ctl->progress - start;
ret = btrfs_remove_free_space(block_group,
start, num_bytes);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ goto out_lock;
start = caching_ctl->progress;
num_bytes = ins->objectid + ins->offset -
caching_ctl->progress;
ret = add_excluded_extent(root, start, num_bytes);
- BUG_ON(ret); /* -ENOMEM */
}
-
+out_lock:
mutex_unlock(&caching_ctl->mutex);
put_caching_control(caching_ctl);
+ if (ret)
+ goto out;
}
ret = btrfs_update_reserved_bytes(block_group, ins->offset,
RESERVE_ALLOC_NO_ACCOUNT);
BUG_ON(ret); /* logic error */
- btrfs_put_block_group(block_group);
ret = alloc_reserved_file_extent(trans, root, 0, root_objectid,
0, owner, offset, ins, 1);
+out:
+ btrfs_put_block_group(block_group);
return ret;
}
-struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- u64 bytenr, u32 blocksize,
- int level)
+static struct extent_buffer *
+btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ u64 bytenr, u32 blocksize, int level)
{
struct extent_buffer *buf;
@@ -6329,12 +6676,14 @@ use_block_rsv(struct btrfs_trans_handle *trans,
if (!ret)
return block_rsv;
if (ret && !block_rsv->failfast) {
- static DEFINE_RATELIMIT_STATE(_rs,
- DEFAULT_RATELIMIT_INTERVAL,
- /*DEFAULT_RATELIMIT_BURST*/ 2);
- if (__ratelimit(&_rs))
- WARN(1, KERN_DEBUG "btrfs: block rsv returned %d\n",
- ret);
+ if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
+ static DEFINE_RATELIMIT_STATE(_rs,
+ DEFAULT_RATELIMIT_INTERVAL * 10,
+ /*DEFAULT_RATELIMIT_BURST*/ 1);
+ if (__ratelimit(&_rs))
+ WARN(1, KERN_DEBUG
+ "btrfs: block rsv returned %d\n", ret);
+ }
ret = reserve_metadata_bytes(root, block_rsv, blocksize,
BTRFS_RESERVE_NO_FLUSH);
if (!ret) {
@@ -6374,7 +6723,8 @@ struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
struct extent_buffer *buf;
u64 flags = 0;
int ret;
-
+ bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
+ SKINNY_METADATA);
block_rsv = use_block_rsv(trans, root, blocksize);
if (IS_ERR(block_rsv))
@@ -6400,14 +6750,17 @@ struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
if (root_objectid != BTRFS_TREE_LOG_OBJECTID) {
struct btrfs_delayed_extent_op *extent_op;
- extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
+ extent_op = btrfs_alloc_delayed_extent_op();
BUG_ON(!extent_op); /* -ENOMEM */
if (key)
memcpy(&extent_op->key, key, sizeof(extent_op->key));
else
memset(&extent_op->key, 0, sizeof(extent_op->key));
extent_op->flags_to_set = flags;
- extent_op->update_key = 1;
+ if (skinny_metadata)
+ extent_op->update_key = 0;
+ else
+ extent_op->update_key = 1;
extent_op->update_flags = 1;
extent_op->is_data = 0;
@@ -6484,8 +6837,9 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
continue;
/* We don't lock the tree block, it's OK to be racy here */
- ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
- &refs, &flags);
+ ret = btrfs_lookup_extent_info(trans, root, bytenr,
+ wc->level - 1, 1, &refs,
+ &flags);
/* We don't care about errors in readahead. */
if (ret < 0)
continue;
@@ -6522,7 +6876,7 @@ reada:
}
/*
- * hepler to process tree block while walking down the tree.
+ * helper to process tree block while walking down the tree.
*
* when wc->stage == UPDATE_BACKREF, this function updates
* back refs for pointers in the block.
@@ -6552,7 +6906,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
(wc->stage == UPDATE_BACKREF && !(wc->flags[level] & flag)))) {
BUG_ON(!path->locks[level]);
ret = btrfs_lookup_extent_info(trans, root,
- eb->start, eb->len,
+ eb->start, level, 1,
&wc->refs[level],
&wc->flags[level]);
BUG_ON(ret == -ENOMEM);
@@ -6597,7 +6951,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
}
/*
- * hepler to process tree block pointer.
+ * helper to process tree block pointer.
*
* when wc->stage == DROP_REFERENCE, this function checks
* reference count of the block pointed to. if the block
@@ -6650,7 +7004,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
- ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
+ ret = btrfs_lookup_extent_info(trans, root, bytenr, level - 1, 1,
&wc->refs[level - 1],
&wc->flags[level - 1]);
if (ret < 0) {
@@ -6658,7 +7012,10 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
return ret;
}
- BUG_ON(wc->refs[level - 1] == 0);
+ if (unlikely(wc->refs[level - 1] == 0)) {
+ btrfs_err(root->fs_info, "Missing references.");
+ BUG();
+ }
*lookup_info = 0;
if (wc->stage == DROP_REFERENCE) {
@@ -6697,8 +7054,10 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
if (reada && level == 1)
reada_walk_down(trans, root, wc, path);
next = read_tree_block(root, bytenr, blocksize, generation);
- if (!next)
+ if (!next || !extent_buffer_uptodate(next)) {
+ free_extent_buffer(next);
return -EIO;
+ }
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
}
@@ -6735,7 +7094,7 @@ skip:
}
/*
- * hepler to process tree block while walking up the tree.
+ * helper to process tree block while walking up the tree.
*
* when wc->stage == DROP_REFERENCE, this function drops
* reference count on the block.
@@ -6781,7 +7140,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
ret = btrfs_lookup_extent_info(trans, root,
- eb->start, eb->len,
+ eb->start, level, 1,
&wc->refs[level],
&wc->flags[level]);
if (ret < 0) {
@@ -6917,6 +7276,8 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
* reference count by one. if update_ref is true, this function
* also make sure backrefs for the shared block and all lower level
* blocks are properly updated.
+ *
+ * If called with for_reloc == 0, may exit early with -EAGAIN
*/
int btrfs_drop_snapshot(struct btrfs_root *root,
struct btrfs_block_rsv *block_rsv, int update_ref,
@@ -6991,8 +7352,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
ret = btrfs_lookup_extent_info(trans, root,
path->nodes[level]->start,
- path->nodes[level]->len,
- &wc->refs[level],
+ level, 1, &wc->refs[level],
&wc->flags[level]);
if (ret < 0) {
err = ret;
@@ -7018,6 +7378,12 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(root);
while (1) {
+ if (!for_reloc && btrfs_fs_closing(root->fs_info)) {
+ pr_debug("btrfs: drop snapshot early exit\n");
+ err = -EAGAIN;
+ goto out_end_trans;
+ }
+
ret = walk_down_tree(trans, root, path, wc);
if (ret < 0) {
err = ret;
@@ -7203,6 +7569,7 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
root->fs_info->fs_devices->missing_devices;
stripped = BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6 |
BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10;
if (num_devices == 1) {
@@ -7481,16 +7848,16 @@ int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
index = get_block_group_index(block_group);
}
- if (index == 0) {
+ if (index == BTRFS_RAID_RAID10) {
dev_min = 4;
/* Divide by 2 */
min_free >>= 1;
- } else if (index == 1) {
+ } else if (index == BTRFS_RAID_RAID1) {
dev_min = 2;
- } else if (index == 2) {
+ } else if (index == BTRFS_RAID_DUP) {
/* Multiply by 2 */
min_free <<= 1;
- } else if (index == 3) {
+ } else if (index == BTRFS_RAID_RAID0) {
dev_min = fs_devices->rw_devices;
do_div(min_free, dev_min);
}
@@ -7651,11 +8018,13 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
space_info = list_entry(info->space_info.next,
struct btrfs_space_info,
list);
- if (space_info->bytes_pinned > 0 ||
- space_info->bytes_reserved > 0 ||
- space_info->bytes_may_use > 0) {
- WARN_ON(1);
- dump_space_info(space_info, 0, 0);
+ if (btrfs_test_opt(info->tree_root, ENOSPC_DEBUG)) {
+ if (space_info->bytes_pinned > 0 ||
+ space_info->bytes_reserved > 0 ||
+ space_info->bytes_may_use > 0) {
+ WARN_ON(1);
+ dump_space_info(space_info, 0, 0);
+ }
}
list_del(&space_info->list);
kfree(space_info);
@@ -7754,7 +8123,9 @@ int btrfs_read_block_groups(struct btrfs_root *root)
btrfs_release_path(path);
cache->flags = btrfs_block_group_flags(&cache->item);
cache->sectorsize = root->sectorsize;
-
+ cache->full_stripe_len = btrfs_full_stripe_len(root,
+ &root->fs_info->mapping_tree,
+ found_key.objectid);
btrfs_init_free_space_ctl(cache);
/*
@@ -7762,7 +8133,17 @@ int btrfs_read_block_groups(struct btrfs_root *root)
* info has super bytes accounted for, otherwise we'll think
* we have more space than we actually do.
*/
- exclude_super_stripes(root, cache);
+ ret = exclude_super_stripes(root, cache);
+ if (ret) {
+ /*
+ * We may have excluded something, so call this just in
+ * case.
+ */
+ free_excluded_extents(root, cache);
+ kfree(cache->free_space_ctl);
+ kfree(cache);
+ goto error;
+ }
/*
* check for two cases, either we are full, and therefore
@@ -7785,10 +8166,26 @@ int btrfs_read_block_groups(struct btrfs_root *root)
free_excluded_extents(root, cache);
}
+ ret = btrfs_add_block_group_cache(root->fs_info, cache);
+ if (ret) {
+ btrfs_remove_free_space_cache(cache);
+ btrfs_put_block_group(cache);
+ goto error;
+ }
+
ret = update_space_info(info, cache->flags, found_key.offset,
btrfs_block_group_used(&cache->item),
&space_info);
- BUG_ON(ret); /* -ENOMEM */
+ 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);
+ spin_unlock(&info->block_group_cache_lock);
+ btrfs_put_block_group(cache);
+ goto error;
+ }
+
cache->space_info = space_info;
spin_lock(&cache->space_info->lock);
cache->space_info->bytes_readonly += cache->bytes_super;
@@ -7796,9 +8193,6 @@ int btrfs_read_block_groups(struct btrfs_root *root)
__link_block_group(space_info, cache);
- ret = btrfs_add_block_group_cache(root->fs_info, cache);
- BUG_ON(ret); /* Logic error */
-
set_avail_alloc_bits(root->fs_info, cache->flags);
if (btrfs_chunk_readonly(root, cache->key.objectid))
set_block_group_ro(cache, 1);
@@ -7808,6 +8202,8 @@ int btrfs_read_block_groups(struct btrfs_root *root)
if (!(get_alloc_profile(root, space_info->flags) &
(BTRFS_BLOCK_GROUP_RAID10 |
BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6 |
BTRFS_BLOCK_GROUP_DUP)))
continue;
/*
@@ -7883,6 +8279,9 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
cache->sectorsize = root->sectorsize;
cache->fs_info = root->fs_info;
+ cache->full_stripe_len = btrfs_full_stripe_len(root,
+ &root->fs_info->mapping_tree,
+ chunk_offset);
atomic_set(&cache->count, 1);
spin_lock_init(&cache->lock);
@@ -7899,16 +8298,41 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
cache->last_byte_to_unpin = (u64)-1;
cache->cached = BTRFS_CACHE_FINISHED;
- exclude_super_stripes(root, cache);
+ ret = exclude_super_stripes(root, cache);
+ if (ret) {
+ /*
+ * We may have excluded something, so call this just in
+ * case.
+ */
+ free_excluded_extents(root, cache);
+ kfree(cache->free_space_ctl);
+ kfree(cache);
+ return ret;
+ }
add_new_free_space(cache, root->fs_info, chunk_offset,
chunk_offset + size);
free_excluded_extents(root, cache);
+ ret = btrfs_add_block_group_cache(root->fs_info, cache);
+ if (ret) {
+ btrfs_remove_free_space_cache(cache);
+ btrfs_put_block_group(cache);
+ return ret;
+ }
+
ret = update_space_info(root->fs_info, cache->flags, size, bytes_used,
&cache->space_info);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret) {
+ btrfs_remove_free_space_cache(cache);
+ spin_lock(&root->fs_info->block_group_cache_lock);
+ rb_erase(&cache->cache_node,
+ &root->fs_info->block_group_cache_tree);
+ spin_unlock(&root->fs_info->block_group_cache_lock);
+ btrfs_put_block_group(cache);
+ return ret;
+ }
update_global_block_rsv(root->fs_info);
spin_lock(&cache->space_info->lock);
@@ -7917,9 +8341,6 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
__link_block_group(cache->space_info, cache);
- ret = btrfs_add_block_group_cache(root->fs_info, cache);
- BUG_ON(ret); /* Logic error */
-
list_add_tail(&cache->new_bg_list, &trans->new_bgs);
set_avail_alloc_bits(extent_root->fs_info, type);
@@ -7932,12 +8353,14 @@ static void clear_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
u64 extra_flags = chunk_to_extended(flags) &
BTRFS_EXTENDED_PROFILE_MASK;
+ write_seqlock(&fs_info->profiles_lock);
if (flags & BTRFS_BLOCK_GROUP_DATA)
fs_info->avail_data_alloc_bits &= ~extra_flags;
if (flags & BTRFS_BLOCK_GROUP_METADATA)
fs_info->avail_metadata_alloc_bits &= ~extra_flags;
if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
fs_info->avail_system_alloc_bits &= ~extra_flags;
+ write_sequnlock(&fs_info->profiles_lock);
}
int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
@@ -8036,6 +8459,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
spin_lock(&root->fs_info->block_group_cache_lock);
rb_erase(&block_group->cache_node,
&root->fs_info->block_group_cache_tree);
+
+ if (root->fs_info->first_logical_byte == block_group->key.objectid)
+ root->fs_info->first_logical_byte = (u64)-1;
spin_unlock(&root->fs_info->block_group_cache_lock);
down_write(&block_group->space_info->groups_sem);
@@ -8158,7 +8584,7 @@ int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range)
if (end - start >= range->minlen) {
if (!block_group_cache_done(cache)) {
- ret = cache_block_group(cache, NULL, root, 0);
+ ret = cache_block_group(cache, 0);
if (!ret)
wait_block_group_cache_done(cache);
}
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 1b319df29ee..32d67a822e9 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -4,7 +4,6 @@
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/page-flags.h>
-#include <linux/module.h>
#include <linux/spinlock.h>
#include <linux/blkdev.h>
#include <linux/swap.h>
@@ -25,12 +24,62 @@
static struct kmem_cache *extent_state_cache;
static struct kmem_cache *extent_buffer_cache;
+#ifdef CONFIG_BTRFS_DEBUG
static LIST_HEAD(buffers);
static LIST_HEAD(states);
-#define LEAK_DEBUG 0
-#if LEAK_DEBUG
static DEFINE_SPINLOCK(leak_lock);
+
+static inline
+void btrfs_leak_debug_add(struct list_head *new, struct list_head *head)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&leak_lock, flags);
+ list_add(new, head);
+ spin_unlock_irqrestore(&leak_lock, flags);
+}
+
+static inline
+void btrfs_leak_debug_del(struct list_head *entry)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&leak_lock, flags);
+ list_del(entry);
+ spin_unlock_irqrestore(&leak_lock, flags);
+}
+
+static inline
+void btrfs_leak_debug_check(void)
+{
+ struct extent_state *state;
+ struct extent_buffer *eb;
+
+ while (!list_empty(&states)) {
+ state = list_entry(states.next, struct extent_state, leak_list);
+ printk(KERN_ERR "btrfs state leak: start %llu end %llu "
+ "state %lu in tree %p refs %d\n",
+ (unsigned long long)state->start,
+ (unsigned long long)state->end,
+ state->state, state->tree, atomic_read(&state->refs));
+ list_del(&state->leak_list);
+ kmem_cache_free(extent_state_cache, state);
+ }
+
+ while (!list_empty(&buffers)) {
+ eb = list_entry(buffers.next, struct extent_buffer, leak_list);
+ printk(KERN_ERR "btrfs buffer leak start %llu len %lu "
+ "refs %d\n", (unsigned long long)eb->start,
+ eb->len, atomic_read(&eb->refs));
+ list_del(&eb->leak_list);
+ kmem_cache_free(extent_buffer_cache, eb);
+ }
+}
+#else
+#define btrfs_leak_debug_add(new, head) do {} while (0)
+#define btrfs_leak_debug_del(entry) do {} while (0)
+#define btrfs_leak_debug_check() do {} while (0)
#endif
#define BUFFER_LRU_MAX 64
@@ -85,29 +134,7 @@ free_state_cache:
void extent_io_exit(void)
{
- struct extent_state *state;
- struct extent_buffer *eb;
-
- while (!list_empty(&states)) {
- state = list_entry(states.next, struct extent_state, leak_list);
- printk(KERN_ERR "btrfs state leak: start %llu end %llu "
- "state %lu in tree %p refs %d\n",
- (unsigned long long)state->start,
- (unsigned long long)state->end,
- state->state, state->tree, atomic_read(&state->refs));
- list_del(&state->leak_list);
- kmem_cache_free(extent_state_cache, state);
-
- }
-
- while (!list_empty(&buffers)) {
- eb = list_entry(buffers.next, struct extent_buffer, leak_list);
- printk(KERN_ERR "btrfs buffer leak start %llu len %lu "
- "refs %d\n", (unsigned long long)eb->start,
- eb->len, atomic_read(&eb->refs));
- list_del(&eb->leak_list);
- kmem_cache_free(extent_buffer_cache, eb);
- }
+ btrfs_leak_debug_check();
/*
* Make sure all delayed rcu free are flushed before we
@@ -135,9 +162,6 @@ void extent_io_tree_init(struct extent_io_tree *tree,
static struct extent_state *alloc_extent_state(gfp_t mask)
{
struct extent_state *state;
-#if LEAK_DEBUG
- unsigned long flags;
-#endif
state = kmem_cache_alloc(extent_state_cache, mask);
if (!state)
@@ -145,11 +169,7 @@ static struct extent_state *alloc_extent_state(gfp_t mask)
state->state = 0;
state->private = 0;
state->tree = NULL;
-#if LEAK_DEBUG
- spin_lock_irqsave(&leak_lock, flags);
- list_add(&state->leak_list, &states);
- spin_unlock_irqrestore(&leak_lock, flags);
-#endif
+ btrfs_leak_debug_add(&state->leak_list, &states);
atomic_set(&state->refs, 1);
init_waitqueue_head(&state->wq);
trace_alloc_extent_state(state, mask, _RET_IP_);
@@ -161,15 +181,8 @@ void free_extent_state(struct extent_state *state)
if (!state)
return;
if (atomic_dec_and_test(&state->refs)) {
-#if LEAK_DEBUG
- unsigned long flags;
-#endif
WARN_ON(state->tree);
-#if LEAK_DEBUG
- spin_lock_irqsave(&leak_lock, flags);
- list_del(&state->leak_list);
- spin_unlock_irqrestore(&leak_lock, flags);
-#endif
+ btrfs_leak_debug_del(&state->leak_list);
trace_free_extent_state(state, _RET_IP_);
kmem_cache_free(extent_state_cache, state);
}
@@ -309,21 +322,21 @@ static void merge_state(struct extent_io_tree *tree,
}
static void set_state_cb(struct extent_io_tree *tree,
- struct extent_state *state, int *bits)
+ struct extent_state *state, unsigned long *bits)
{
if (tree->ops && tree->ops->set_bit_hook)
tree->ops->set_bit_hook(tree->mapping->host, state, bits);
}
static void clear_state_cb(struct extent_io_tree *tree,
- struct extent_state *state, int *bits)
+ struct extent_state *state, unsigned long *bits)
{
if (tree->ops && tree->ops->clear_bit_hook)
tree->ops->clear_bit_hook(tree->mapping->host, state, bits);
}
static void set_state_bits(struct extent_io_tree *tree,
- struct extent_state *state, int *bits);
+ struct extent_state *state, unsigned long *bits);
/*
* insert an extent_state struct into the tree. 'bits' are set on the
@@ -337,7 +350,7 @@ static void set_state_bits(struct extent_io_tree *tree,
*/
static int insert_state(struct extent_io_tree *tree,
struct extent_state *state, u64 start, u64 end,
- int *bits)
+ unsigned long *bits)
{
struct rb_node *node;
@@ -425,10 +438,10 @@ static struct extent_state *next_state(struct extent_state *state)
*/
static struct extent_state *clear_state_bit(struct extent_io_tree *tree,
struct extent_state *state,
- int *bits, int wake)
+ unsigned long *bits, int wake)
{
struct extent_state *next;
- int bits_to_clear = *bits & ~EXTENT_CTLBITS;
+ unsigned long bits_to_clear = *bits & ~EXTENT_CTLBITS;
if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
u64 range = state->end - state->start + 1;
@@ -464,7 +477,7 @@ alloc_extent_state_atomic(struct extent_state *prealloc)
return prealloc;
}
-void extent_io_tree_panic(struct extent_io_tree *tree, int err)
+static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
{
btrfs_panic(tree_fs_info(tree), err, "Locking error: "
"Extent tree was modified by another "
@@ -484,7 +497,7 @@ void extent_io_tree_panic(struct extent_io_tree *tree, int err)
* This takes the tree lock, and returns 0 on success and < 0 on error.
*/
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int wake, int delete,
+ unsigned long bits, int wake, int delete,
struct extent_state **cached_state,
gfp_t mask)
{
@@ -645,7 +658,8 @@ static void wait_on_state(struct extent_io_tree *tree,
* The range [start, end] is inclusive.
* The tree lock is taken by this function
*/
-void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits)
+static void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+ unsigned long bits)
{
struct extent_state *state;
struct rb_node *node;
@@ -686,9 +700,9 @@ out:
static void set_state_bits(struct extent_io_tree *tree,
struct extent_state *state,
- int *bits)
+ unsigned long *bits)
{
- int bits_to_set = *bits & ~EXTENT_CTLBITS;
+ unsigned long bits_to_set = *bits & ~EXTENT_CTLBITS;
set_state_cb(tree, state, bits);
if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) {
@@ -731,8 +745,9 @@ static void uncache_state(struct extent_state **cached_ptr)
static int __must_check
__set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int exclusive_bits, u64 *failed_start,
- struct extent_state **cached_state, gfp_t mask)
+ unsigned long bits, unsigned long exclusive_bits,
+ u64 *failed_start, struct extent_state **cached_state,
+ gfp_t mask)
{
struct extent_state *state;
struct extent_state *prealloc = NULL;
@@ -924,9 +939,9 @@ search_again:
goto again;
}
-int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits,
- u64 *failed_start, struct extent_state **cached_state,
- gfp_t mask)
+int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+ unsigned long bits, u64 * failed_start,
+ struct extent_state **cached_state, gfp_t mask)
{
return __set_extent_bit(tree, start, end, bits, 0, failed_start,
cached_state, mask);
@@ -951,7 +966,7 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits,
* boundary bits like LOCK.
*/
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int clear_bits,
+ unsigned long bits, unsigned long clear_bits,
struct extent_state **cached_state, gfp_t mask)
{
struct extent_state *state;
@@ -1144,14 +1159,14 @@ int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
}
int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, gfp_t mask)
+ unsigned long bits, gfp_t mask)
{
return set_extent_bit(tree, start, end, bits, NULL,
NULL, mask);
}
int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, gfp_t mask)
+ unsigned long bits, gfp_t mask)
{
return clear_extent_bit(tree, start, end, bits, 0, 0, NULL, mask);
}
@@ -1190,7 +1205,7 @@ int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask)
{
- return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0,
+ return set_extent_bit(tree, start, end, EXTENT_UPTODATE, NULL,
cached_state, mask);
}
@@ -1206,7 +1221,7 @@ int clear_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
* us if waiting is desired.
*/
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, struct extent_state **cached_state)
+ unsigned long bits, struct extent_state **cached_state)
{
int err;
u64 failed_start;
@@ -1258,6 +1273,39 @@ int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
GFP_NOFS);
}
+int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end)
+{
+ unsigned long index = start >> PAGE_CACHE_SHIFT;
+ unsigned long end_index = end >> PAGE_CACHE_SHIFT;
+ struct page *page;
+
+ while (index <= end_index) {
+ page = find_get_page(inode->i_mapping, index);
+ BUG_ON(!page); /* Pages should be in the extent_io_tree */
+ clear_page_dirty_for_io(page);
+ page_cache_release(page);
+ index++;
+ }
+ return 0;
+}
+
+int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end)
+{
+ unsigned long index = start >> PAGE_CACHE_SHIFT;
+ unsigned long end_index = end >> PAGE_CACHE_SHIFT;
+ struct page *page;
+
+ while (index <= end_index) {
+ page = find_get_page(inode->i_mapping, index);
+ BUG_ON(!page); /* Pages should be in the extent_io_tree */
+ account_page_redirty(page);
+ __set_page_dirty_nobuffers(page);
+ page_cache_release(page);
+ index++;
+ }
+ return 0;
+}
+
/*
* helper function to set both pages and extents in the tree writeback
*/
@@ -1281,8 +1329,9 @@ static int set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
* return it. tree->lock must be held. NULL will returned if
* nothing was found after 'start'
*/
-struct extent_state *find_first_extent_bit_state(struct extent_io_tree *tree,
- u64 start, int bits)
+static struct extent_state *
+find_first_extent_bit_state(struct extent_io_tree *tree,
+ u64 start, unsigned long bits)
{
struct rb_node *node;
struct extent_state *state;
@@ -1316,7 +1365,7 @@ out:
* If nothing was found, 1 is returned. If found something, return 0.
*/
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
- u64 *start_ret, u64 *end_ret, int bits,
+ u64 *start_ret, u64 *end_ret, unsigned long bits,
struct extent_state **cached_state)
{
struct extent_state *state;
@@ -1606,7 +1655,7 @@ int extent_clear_unlock_delalloc(struct inode *inode,
unsigned long end_index = end >> PAGE_CACHE_SHIFT;
unsigned long nr_pages = end_index - index + 1;
int i;
- int clear_bits = 0;
+ unsigned long clear_bits = 0;
if (op & EXTENT_CLEAR_UNLOCK)
clear_bits |= EXTENT_LOCKED;
@@ -1745,6 +1794,64 @@ out:
return ret;
}
+void extent_cache_csums_dio(struct extent_io_tree *tree, u64 start, u32 csums[],
+ int count)
+{
+ struct rb_node *node;
+ struct extent_state *state;
+
+ spin_lock(&tree->lock);
+ /*
+ * this search will find all the extents that end after
+ * our range starts.
+ */
+ node = tree_search(tree, start);
+ BUG_ON(!node);
+
+ state = rb_entry(node, struct extent_state, rb_node);
+ BUG_ON(state->start != start);
+
+ while (count) {
+ state->private = *csums++;
+ count--;
+ state = next_state(state);
+ }
+ spin_unlock(&tree->lock);
+}
+
+static inline u64 __btrfs_get_bio_offset(struct bio *bio, int bio_index)
+{
+ struct bio_vec *bvec = bio->bi_io_vec + bio_index;
+
+ return page_offset(bvec->bv_page) + bvec->bv_offset;
+}
+
+void extent_cache_csums(struct extent_io_tree *tree, struct bio *bio, int bio_index,
+ u32 csums[], int count)
+{
+ struct rb_node *node;
+ struct extent_state *state = NULL;
+ u64 start;
+
+ spin_lock(&tree->lock);
+ do {
+ start = __btrfs_get_bio_offset(bio, bio_index);
+ if (state == NULL || state->start != start) {
+ node = tree_search(tree, start);
+ BUG_ON(!node);
+
+ state = rb_entry(node, struct extent_state, rb_node);
+ BUG_ON(state->start != start);
+ }
+ state->private = *csums++;
+ count--;
+ bio_index++;
+
+ state = next_state(state);
+ } while (count);
+ spin_unlock(&tree->lock);
+}
+
int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private)
{
struct rb_node *node;
@@ -1779,7 +1886,7 @@ out:
* range is found set.
*/
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int filled, struct extent_state *cached)
+ unsigned long bits, int filled, struct extent_state *cached)
{
struct extent_state *state = NULL;
struct rb_node *node;
@@ -1834,7 +1941,7 @@ int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
*/
static void check_page_uptodate(struct extent_io_tree *tree, struct page *page)
{
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 end = start + PAGE_CACHE_SIZE - 1;
if (test_range_bit(tree, start, end, EXTENT_UPTODATE, 1, NULL))
SetPageUptodate(page);
@@ -1846,7 +1953,7 @@ static void check_page_uptodate(struct extent_io_tree *tree, struct page *page)
*/
static void check_page_locked(struct extent_io_tree *tree, struct page *page)
{
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 end = start + PAGE_CACHE_SIZE - 1;
if (!test_range_bit(tree, start, end, EXTENT_LOCKED, 0, NULL))
unlock_page(page);
@@ -1895,13 +2002,11 @@ static int free_io_failure(struct inode *inode, struct io_failure_record *rec,
if (ret)
err = ret;
- if (did_repair) {
- ret = clear_extent_bits(&BTRFS_I(inode)->io_tree, rec->start,
- rec->start + rec->len - 1,
- EXTENT_DAMAGED, GFP_NOFS);
- if (ret && !err)
- err = ret;
- }
+ ret = clear_extent_bits(&BTRFS_I(inode)->io_tree, rec->start,
+ rec->start + rec->len - 1,
+ EXTENT_DAMAGED, GFP_NOFS);
+ if (ret && !err)
+ err = ret;
kfree(rec);
return err;
@@ -1932,10 +2037,15 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 start,
u64 map_length = 0;
u64 sector;
struct btrfs_bio *bbio = NULL;
+ struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
int ret;
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 = bio_alloc(GFP_NOFS, 1);
if (!bio)
return -EIO;
@@ -1960,7 +2070,7 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 start,
return -EIO;
}
bio->bi_bdev = dev->bdev;
- bio_add_page(bio, page, length, start-page_offset(page));
+ bio_add_page(bio, page, length, start - page_offset(page));
btrfsic_submit_bio(WRITE_SYNC, bio);
wait_for_completion(&compl);
@@ -2052,6 +2162,7 @@ static int clean_io_failure(u64 start, struct page *page)
failrec->failed_mirror);
did_repair = !ret;
}
+ ret = 0;
}
out:
@@ -2293,8 +2404,7 @@ static void end_bio_extent_writepage(struct bio *bio, int err)
struct page *page = bvec->bv_page;
tree = &BTRFS_I(page->mapping->host)->io_tree;
- start = ((u64)page->index << PAGE_CACHE_SHIFT) +
- bvec->bv_offset;
+ start = page_offset(page) + bvec->bv_offset;
end = start + bvec->bv_len - 1;
if (bvec->bv_offset == 0 && bvec->bv_len == PAGE_CACHE_SIZE)
@@ -2353,8 +2463,7 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
(long int)bio->bi_bdev);
tree = &BTRFS_I(page->mapping->host)->io_tree;
- start = ((u64)page->index << PAGE_CACHE_SHIFT) +
- bvec->bv_offset;
+ start = page_offset(page) + bvec->bv_offset;
end = start + bvec->bv_len - 1;
if (bvec->bv_offset == 0 && bvec->bv_len == PAGE_CACHE_SIZE)
@@ -2471,7 +2580,7 @@ static int __must_check submit_one_bio(int rw, struct bio *bio,
struct extent_io_tree *tree = bio->bi_private;
u64 start;
- start = ((u64)page->index << PAGE_CACHE_SHIFT) + bvec->bv_offset;
+ start = page_offset(page) + bvec->bv_offset;
bio->bi_private = NULL;
@@ -2489,13 +2598,13 @@ static int __must_check submit_one_bio(int rw, struct bio *bio,
return ret;
}
-static int merge_bio(struct extent_io_tree *tree, struct page *page,
+static int merge_bio(int rw, struct extent_io_tree *tree, struct page *page,
unsigned long offset, size_t size, struct bio *bio,
unsigned long bio_flags)
{
int ret = 0;
if (tree->ops && tree->ops->merge_bio_hook)
- ret = tree->ops->merge_bio_hook(page, offset, size, bio,
+ ret = tree->ops->merge_bio_hook(rw, page, offset, size, bio,
bio_flags);
BUG_ON(ret < 0);
return ret;
@@ -2526,11 +2635,10 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
if (old_compressed)
contig = bio->bi_sector == sector;
else
- contig = bio->bi_sector + (bio->bi_size >> 9) ==
- sector;
+ contig = bio_end_sector(bio) == sector;
if (prev_bio_flags != bio_flags || !contig ||
- merge_bio(tree, page, offset, page_size, bio, bio_flags) ||
+ merge_bio(rw, tree, page, offset, page_size, bio, bio_flags) ||
bio_add_page(bio, page, page_size, offset) < page_size) {
ret = submit_one_bio(rw, bio, mirror_num,
prev_bio_flags);
@@ -2562,7 +2670,8 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
return ret;
}
-void attach_extent_buffer_page(struct extent_buffer *eb, struct page *page)
+static void attach_extent_buffer_page(struct extent_buffer *eb,
+ struct page *page)
{
if (!PagePrivate(page)) {
SetPagePrivate(page);
@@ -2592,10 +2701,10 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
struct page *page,
get_extent_t *get_extent,
struct bio **bio, int mirror_num,
- unsigned long *bio_flags)
+ unsigned long *bio_flags, int rw)
{
struct inode *inode = page->mapping->host;
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 page_end = start + PAGE_CACHE_SIZE - 1;
u64 end;
u64 cur = start;
@@ -2648,6 +2757,8 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
}
}
while (cur <= end) {
+ unsigned long pnr = (last_byte >> PAGE_CACHE_SHIFT) + 1;
+
if (cur >= last_byte) {
char *userpage;
struct extent_state *cached = NULL;
@@ -2682,7 +2793,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
iosize = min(extent_map_end(em) - cur, end - cur + 1);
cur_end = min(extent_map_end(em) - 1, end);
- iosize = (iosize + blocksize - 1) & ~((u64)blocksize - 1);
+ iosize = ALIGN(iosize, blocksize);
if (this_bio_flag & EXTENT_BIO_COMPRESSED) {
disk_io_size = em->block_len;
sector = em->block_start >> 9;
@@ -2735,26 +2846,17 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
continue;
}
- ret = 0;
- if (tree->ops && tree->ops->readpage_io_hook) {
- ret = tree->ops->readpage_io_hook(page, cur,
- cur + iosize - 1);
- }
- if (!ret) {
- unsigned long pnr = (last_byte >> PAGE_CACHE_SHIFT) + 1;
- pnr -= page->index;
- ret = submit_extent_page(READ, tree, page,
+ pnr -= page->index;
+ ret = submit_extent_page(rw, tree, page,
sector, disk_io_size, pg_offset,
bdev, bio, pnr,
end_bio_extent_readpage, mirror_num,
*bio_flags,
this_bio_flag);
- if (!ret) {
- nr++;
- *bio_flags = this_bio_flag;
- }
- }
- if (ret) {
+ if (!ret) {
+ nr++;
+ *bio_flags = this_bio_flag;
+ } else {
SetPageError(page);
unlock_extent(tree, cur, cur + iosize - 1);
}
@@ -2778,7 +2880,7 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
int ret;
ret = __extent_read_full_page(tree, page, get_extent, &bio, mirror_num,
- &bio_flags);
+ &bio_flags, READ);
if (bio)
ret = submit_one_bio(READ, bio, mirror_num, bio_flags);
return ret;
@@ -2806,7 +2908,7 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
struct inode *inode = page->mapping->host;
struct extent_page_data *epd = data;
struct extent_io_tree *tree = epd->tree;
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 delalloc_start;
u64 page_end = start + PAGE_CACHE_SIZE - 1;
u64 end;
@@ -2982,7 +3084,7 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
BUG_ON(extent_map_end(em) <= cur);
BUG_ON(end < cur);
iosize = min(extent_map_end(em) - cur, end - cur + 1);
- iosize = (iosize + blocksize - 1) & ~((u64)blocksize - 1);
+ iosize = ALIGN(iosize, blocksize);
sector = (em->block_start + extent_offset) >> 9;
bdev = em->bdev;
block_start = em->block_start;
@@ -3077,7 +3179,7 @@ static int eb_wait(void *word)
return 0;
}
-static void wait_on_extent_buffer_writeback(struct extent_buffer *eb)
+void wait_on_extent_buffer_writeback(struct extent_buffer *eb)
{
wait_on_bit(&eb->bflags, EXTENT_BUFFER_WRITEBACK, eb_wait,
TASK_UNINTERRUPTIBLE);
@@ -3124,12 +3226,9 @@ static int 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);
- spin_lock(&fs_info->delalloc_lock);
- if (fs_info->dirty_metadata_bytes >= eb->len)
- fs_info->dirty_metadata_bytes -= eb->len;
- else
- WARN_ON(1);
- spin_unlock(&fs_info->delalloc_lock);
+ __percpu_counter_add(&fs_info->dirty_metadata_bytes,
+ -eb->len,
+ fs_info->dirty_metadata_batch);
ret = 1;
} else {
spin_unlock(&eb->refs_lock);
@@ -3205,7 +3304,7 @@ static int write_one_eb(struct extent_buffer *eb,
u64 offset = eb->start;
unsigned long i, num_pages;
unsigned long bio_flags = 0;
- int rw = (epd->sync_io ? WRITE_SYNC : WRITE);
+ int rw = (epd->sync_io ? WRITE_SYNC : WRITE) | REQ_META;
int ret = 0;
clear_bit(EXTENT_BUFFER_IOERR, &eb->bflags);
@@ -3446,15 +3545,9 @@ retry:
* swizzled back from swapper_space to tmpfs file
* mapping
*/
- if (tree->ops &&
- tree->ops->write_cache_pages_lock_hook) {
- tree->ops->write_cache_pages_lock_hook(page,
- data, flush_fn);
- } else {
- if (!trylock_page(page)) {
- flush_fn(data);
- lock_page(page);
- }
+ if (!trylock_page(page)) {
+ flush_fn(data);
+ lock_page(page);
}
if (unlikely(page->mapping != mapping)) {
@@ -3648,14 +3741,14 @@ int extent_readpages(struct extent_io_tree *tree,
continue;
for (i = 0; i < nr; i++) {
__extent_read_full_page(tree, pagepool[i], get_extent,
- &bio, 0, &bio_flags);
+ &bio, 0, &bio_flags, READ);
page_cache_release(pagepool[i]);
}
nr = 0;
}
for (i = 0; i < nr; i++) {
__extent_read_full_page(tree, pagepool[i], get_extent,
- &bio, 0, &bio_flags);
+ &bio, 0, &bio_flags, READ);
page_cache_release(pagepool[i]);
}
@@ -3674,11 +3767,11 @@ int extent_invalidatepage(struct extent_io_tree *tree,
struct page *page, unsigned long offset)
{
struct extent_state *cached_state = NULL;
- u64 start = ((u64)page->index << PAGE_CACHE_SHIFT);
+ u64 start = page_offset(page);
u64 end = start + PAGE_CACHE_SIZE - 1;
size_t blocksize = page->mapping->host->i_sb->s_blocksize;
- start += (offset + blocksize - 1) & ~(blocksize - 1);
+ start += ALIGN(offset, blocksize);
if (start > end)
return 0;
@@ -3696,11 +3789,11 @@ int extent_invalidatepage(struct extent_io_tree *tree,
* are locked or under IO and drops the related state bits if it is safe
* to drop the page.
*/
-int try_release_extent_state(struct extent_map_tree *map,
- struct extent_io_tree *tree, struct page *page,
- gfp_t mask)
+static int try_release_extent_state(struct extent_map_tree *map,
+ struct extent_io_tree *tree,
+ struct page *page, gfp_t mask)
{
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 end = start + PAGE_CACHE_SIZE - 1;
int ret = 1;
@@ -3739,7 +3832,7 @@ int try_release_extent_mapping(struct extent_map_tree *map,
gfp_t mask)
{
struct extent_map *em;
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
+ u64 start = page_offset(page);
u64 end = start + PAGE_CACHE_SIZE - 1;
if ((mask & __GFP_WAIT) &&
@@ -3797,7 +3890,7 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
len = last - offset;
if (len == 0)
break;
- len = (len + sectorsize - 1) & ~(sectorsize - 1);
+ len = ALIGN(len, sectorsize);
em = get_extent(inode, NULL, 0, offset, len, 0);
if (IS_ERR_OR_NULL(em))
return em;
@@ -3989,14 +4082,7 @@ out:
static void __free_extent_buffer(struct extent_buffer *eb)
{
-#if LEAK_DEBUG
- unsigned long flags;
- spin_lock_irqsave(&leak_lock, flags);
- list_del(&eb->leak_list);
- spin_unlock_irqrestore(&leak_lock, flags);
-#endif
- if (eb->pages && eb->pages != eb->inline_pages)
- kfree(eb->pages);
+ btrfs_leak_debug_del(&eb->leak_list);
kmem_cache_free(extent_buffer_cache, eb);
}
@@ -4006,9 +4092,6 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree,
gfp_t mask)
{
struct extent_buffer *eb = NULL;
-#if LEAK_DEBUG
- unsigned long flags;
-#endif
eb = kmem_cache_zalloc(extent_buffer_cache, mask);
if (eb == NULL)
@@ -4028,28 +4111,18 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree,
init_waitqueue_head(&eb->write_lock_wq);
init_waitqueue_head(&eb->read_lock_wq);
-#if LEAK_DEBUG
- spin_lock_irqsave(&leak_lock, flags);
- list_add(&eb->leak_list, &buffers);
- spin_unlock_irqrestore(&leak_lock, flags);
-#endif
+ btrfs_leak_debug_add(&eb->leak_list, &buffers);
+
spin_lock_init(&eb->refs_lock);
atomic_set(&eb->refs, 1);
atomic_set(&eb->io_pages, 0);
- if (len > MAX_INLINE_EXTENT_BUFFER_SIZE) {
- struct page **pages;
- int num_pages = (len + PAGE_CACHE_SIZE - 1) >>
- PAGE_CACHE_SHIFT;
- pages = kzalloc(num_pages, mask);
- if (!pages) {
- __free_extent_buffer(eb);
- return NULL;
- }
- eb->pages = pages;
- } else {
- eb->pages = eb->inline_pages;
- }
+ /*
+ * Sanity checks, currently the maximum is 64k covered by 16x 4k pages
+ */
+ BUILD_BUG_ON(BTRFS_MAX_METADATA_BLOCKSIZE
+ > MAX_INLINE_EXTENT_BUFFER_SIZE);
+ BUG_ON(len > MAX_INLINE_EXTENT_BUFFER_SIZE);
return eb;
}
@@ -4180,6 +4253,7 @@ static inline void btrfs_release_extent_buffer(struct extent_buffer *eb)
static void check_buffer_tree_ref(struct extent_buffer *eb)
{
+ int refs;
/* the ref bit is tricky. We have to make sure it is set
* if we have the buffer dirty. Otherwise the
* code to free a buffer can end up dropping a dirty
@@ -4200,6 +4274,10 @@ static void check_buffer_tree_ref(struct extent_buffer *eb)
* So bump the ref count first, then set the bit. If someone
* beat us to it, drop the ref we added.
*/
+ refs = atomic_read(&eb->refs);
+ if (refs >= 2 && test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags))
+ return;
+
spin_lock(&eb->refs_lock);
if (!test_and_set_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags))
atomic_inc(&eb->refs);
@@ -4372,7 +4450,7 @@ static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head)
}
/* Expects to have eb->eb_lock already held */
-static int release_extent_buffer(struct extent_buffer *eb, gfp_t mask)
+static int release_extent_buffer(struct extent_buffer *eb)
{
WARN_ON(atomic_read(&eb->refs) == 0);
if (atomic_dec_and_test(&eb->refs)) {
@@ -4401,9 +4479,20 @@ static int release_extent_buffer(struct extent_buffer *eb, gfp_t mask)
void free_extent_buffer(struct extent_buffer *eb)
{
+ int refs;
+ int old;
if (!eb)
return;
+ while (1) {
+ refs = atomic_read(&eb->refs);
+ if (refs <= 3)
+ break;
+ old = atomic_cmpxchg(&eb->refs, refs, refs - 1);
+ if (old == refs)
+ return;
+ }
+
spin_lock(&eb->refs_lock);
if (atomic_read(&eb->refs) == 2 &&
test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags))
@@ -4419,7 +4508,7 @@ void free_extent_buffer(struct extent_buffer *eb)
* I know this is terrible, but it's temporary until we stop tracking
* the uptodate bits and such for the extent buffers.
*/
- release_extent_buffer(eb, GFP_ATOMIC);
+ release_extent_buffer(eb);
}
void free_extent_buffer_stale(struct extent_buffer *eb)
@@ -4433,7 +4522,7 @@ void free_extent_buffer_stale(struct extent_buffer *eb)
if (atomic_read(&eb->refs) == 2 && !extent_buffer_under_io(eb) &&
test_and_clear_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags))
atomic_dec(&eb->refs);
- release_extent_buffer(eb, GFP_NOFS);
+ release_extent_buffer(eb);
}
void clear_extent_buffer_dirty(struct extent_buffer *eb)
@@ -4485,17 +4574,6 @@ int set_extent_buffer_dirty(struct extent_buffer *eb)
return was_dirty;
}
-static int range_straddles_pages(u64 start, u64 len)
-{
- if (len < PAGE_CACHE_SIZE)
- return 1;
- if (start & (PAGE_CACHE_SIZE - 1))
- return 1;
- if ((start + len) & (PAGE_CACHE_SIZE - 1))
- return 1;
- return 0;
-}
-
int clear_extent_buffer_uptodate(struct extent_buffer *eb)
{
unsigned long i;
@@ -4527,37 +4605,6 @@ int set_extent_buffer_uptodate(struct extent_buffer *eb)
return 0;
}
-int extent_range_uptodate(struct extent_io_tree *tree,
- u64 start, u64 end)
-{
- struct page *page;
- int ret;
- int pg_uptodate = 1;
- int uptodate;
- unsigned long index;
-
- if (range_straddles_pages(start, end - start + 1)) {
- ret = test_range_bit(tree, start, end,
- EXTENT_UPTODATE, 1, NULL);
- if (ret)
- return 1;
- }
- while (start <= end) {
- index = start >> PAGE_CACHE_SHIFT;
- page = find_get_page(tree->mapping, index);
- if (!page)
- return 1;
- uptodate = PageUptodate(page);
- page_cache_release(page);
- if (!uptodate) {
- pg_uptodate = 0;
- break;
- }
- start += PAGE_CACHE_SIZE;
- }
- return pg_uptodate;
-}
-
int extent_buffer_uptodate(struct extent_buffer *eb)
{
return test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
@@ -4620,7 +4667,8 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
ClearPageError(page);
err = __extent_read_full_page(tree, page,
get_extent, &bio,
- mirror_num, &bio_flags);
+ mirror_num, &bio_flags,
+ READ | REQ_META);
if (err)
ret = err;
} else {
@@ -4629,7 +4677,8 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
}
if (bio) {
- err = submit_one_bio(READ, bio, mirror_num, bio_flags);
+ err = submit_one_bio(READ | REQ_META, bio, mirror_num,
+ bio_flags);
if (err)
return err;
}
@@ -4993,7 +5042,7 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
}
}
-int try_release_extent_buffer(struct page *page, gfp_t mask)
+int try_release_extent_buffer(struct page *page)
{
struct extent_buffer *eb;
@@ -5023,9 +5072,6 @@ int try_release_extent_buffer(struct page *page, gfp_t mask)
}
spin_unlock(&page->mapping->private_lock);
- if ((mask & GFP_NOFS) == GFP_NOFS)
- mask = GFP_NOFS;
-
/*
* If tree ref isn't set then we know the ref on this eb is a real ref,
* so just return, this page will likely be freed soon anyway.
@@ -5035,5 +5081,5 @@ int try_release_extent_buffer(struct page *page, gfp_t mask)
return 0;
}
- return release_extent_buffer(eb, mask);
+ return release_extent_buffer(eb);
}
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 2eacfabd326..a2c03a17500 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -72,26 +72,23 @@ struct extent_io_ops {
int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
int (*writepage_io_hook)(struct page *page, u64 start, u64 end);
extent_submit_bio_hook_t *submit_bio_hook;
- int (*merge_bio_hook)(struct page *page, unsigned long offset,
+ int (*merge_bio_hook)(int rw, struct page *page, unsigned long offset,
size_t size, struct bio *bio,
unsigned long bio_flags);
- int (*readpage_io_hook)(struct page *page, u64 start, u64 end);
int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
int (*readpage_end_io_hook)(struct page *page, u64 start, u64 end,
struct extent_state *state, int mirror);
int (*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,
- int *bits);
+ unsigned long *bits);
void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
- int *bits);
+ unsigned long *bits);
void (*merge_extent_hook)(struct inode *inode,
struct extent_state *new,
struct extent_state *other);
void (*split_extent_hook)(struct inode *inode,
struct extent_state *orig, u64 split);
- int (*write_cache_pages_lock_hook)(struct page *page, void *data,
- void (*flush_fn)(void *));
};
struct extent_io_tree {
@@ -119,7 +116,9 @@ struct extent_state {
/* for use by the FS */
u64 private;
+#ifdef CONFIG_BTRFS_DEBUG
struct list_head leak_list;
+#endif
};
#define INLINE_EXTENT_BUFFER_PAGES 16
@@ -135,7 +134,6 @@ struct extent_buffer {
atomic_t refs;
atomic_t io_pages;
int read_mirror;
- struct list_head leak_list;
struct rcu_head rcu_head;
pid_t lock_owner;
@@ -161,8 +159,10 @@ struct extent_buffer {
*/
wait_queue_head_t read_lock_wq;
wait_queue_head_t lock_wq;
- struct page *inline_pages[INLINE_EXTENT_BUFFER_PAGES];
- struct page **pages;
+ struct page *pages[INLINE_EXTENT_BUFFER_PAGES];
+#ifdef CONFIG_BTRFS_DEBUG
+ struct list_head leak_list;
+#endif
};
static inline void extent_set_compress_type(unsigned long *bio_flags,
@@ -189,13 +189,10 @@ void extent_io_tree_init(struct extent_io_tree *tree,
int try_release_extent_mapping(struct extent_map_tree *map,
struct extent_io_tree *tree, struct page *page,
gfp_t mask);
-int try_release_extent_buffer(struct page *page, gfp_t mask);
-int try_release_extent_state(struct extent_map_tree *map,
- struct extent_io_tree *tree, struct page *page,
- gfp_t mask);
+int try_release_extent_buffer(struct page *page);
int lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, struct extent_state **cached);
+ unsigned long bits, struct extent_state **cached);
int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end);
int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached, gfp_t mask);
@@ -211,16 +208,17 @@ u64 count_range_bits(struct extent_io_tree *tree,
void free_extent_state(struct extent_state *state);
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int filled, struct extent_state *cached_state);
+ unsigned long bits, int filled,
+ struct extent_state *cached_state);
int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, gfp_t mask);
+ unsigned long bits, gfp_t mask);
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int wake, int delete, struct extent_state **cached,
- gfp_t mask);
+ unsigned long bits, int wake, int delete,
+ struct extent_state **cached, gfp_t mask);
int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, gfp_t mask);
+ unsigned long bits, gfp_t mask);
int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, u64 *failed_start,
+ unsigned long bits, u64 *failed_start,
struct extent_state **cached_state, gfp_t mask);
int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
@@ -233,17 +231,15 @@ int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask);
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int clear_bits,
+ unsigned long bits, unsigned long clear_bits,
struct extent_state **cached_state, gfp_t mask);
int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end,
struct extent_state **cached_state, gfp_t mask);
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
- u64 *start_ret, u64 *end_ret, int bits,
+ u64 *start_ret, u64 *end_ret, unsigned long bits,
struct extent_state **cached_state);
-struct extent_state *find_first_extent_bit_state(struct extent_io_tree *tree,
- u64 start, int bits);
int extent_invalidatepage(struct extent_io_tree *tree,
struct page *page, unsigned long offset);
int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
@@ -265,6 +261,10 @@ int extent_readpages(struct extent_io_tree *tree,
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
__u64 start, __u64 len, get_extent_t *get_extent);
int set_state_private(struct extent_io_tree *tree, u64 start, u64 private);
+void extent_cache_csums_dio(struct extent_io_tree *tree, u64 start, u32 csums[],
+ int count);
+void extent_cache_csums(struct extent_io_tree *tree, struct bio *bio,
+ int bvec_index, u32 csums[], int count);
int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private);
void set_page_extent_mapped(struct page *page);
@@ -282,6 +282,7 @@ void free_extent_buffer_stale(struct extent_buffer *eb);
int read_extent_buffer_pages(struct extent_io_tree *tree,
struct extent_buffer *eb, u64 start, int wait,
get_extent_t *get_extent, int mirror_num);
+void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
static inline unsigned long num_extent_pages(u64 start, u64 len)
{
@@ -317,7 +318,6 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
unsigned long src_offset, unsigned long len);
void memset_extent_buffer(struct extent_buffer *eb, char c,
unsigned long start, unsigned long len);
-void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits);
void clear_extent_buffer_dirty(struct extent_buffer *eb);
int set_extent_buffer_dirty(struct extent_buffer *eb);
int set_extent_buffer_uptodate(struct extent_buffer *eb);
@@ -327,8 +327,8 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
unsigned long min_len, char **map,
unsigned long *map_start,
unsigned long *map_len);
-int extent_range_uptodate(struct extent_io_tree *tree,
- u64 start, u64 end);
+int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
+int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
int extent_clear_unlock_delalloc(struct inode *inode,
struct extent_io_tree *tree,
u64 start, u64 end, struct page *locked_page,
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index fdb7a8db3b5..a4a7a1a8da9 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -1,6 +1,5 @@
#include <linux/err.h>
#include <linux/slab.h>
-#include <linux/module.h>
#include <linux/spinlock.h>
#include <linux/hardirq.h>
#include "ctree.h"
@@ -175,6 +174,14 @@ static int mergable_maps(struct extent_map *prev, struct extent_map *next)
test_bit(EXTENT_FLAG_LOGGING, &next->flags))
return 0;
+ /*
+ * We don't want to merge stuff that hasn't been written to the log yet
+ * since it may not reflect exactly what is on disk, and that would be
+ * bad.
+ */
+ if (!list_empty(&prev->list) || !list_empty(&next->list))
+ return 0;
+
if (extent_map_end(prev) == next->start &&
prev->flags == next->flags &&
prev->bdev == next->bdev &&
@@ -210,9 +217,7 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
em->mod_len = (em->mod_len + em->mod_start) - merge->mod_start;
em->mod_start = merge->mod_start;
em->generation = max(em->generation, merge->generation);
- list_move(&em->list, &tree->modified_extents);
- list_del_init(&merge->list);
rb_erase(&merge->rb_node, &tree->map);
free_extent_map(merge);
}
@@ -228,7 +233,6 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
merge->in_tree = 0;
em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start;
em->generation = max(em->generation, merge->generation);
- list_del_init(&merge->list);
free_extent_map(merge);
}
}
@@ -303,7 +307,7 @@ void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em)
* reference dropped if the merge attempt was successful.
*/
int add_extent_mapping(struct extent_map_tree *tree,
- struct extent_map *em)
+ struct extent_map *em, int modified)
{
int ret = 0;
struct rb_node *rb;
@@ -325,7 +329,10 @@ int add_extent_mapping(struct extent_map_tree *tree,
em->mod_start = em->start;
em->mod_len = em->len;
- try_merge_map(tree, em);
+ if (modified)
+ list_move(&em->list, &tree->modified_extents);
+ else
+ try_merge_map(tree, em);
out:
return ret;
}
@@ -338,8 +345,9 @@ static u64 range_end(u64 start, u64 len)
return start + len;
}
-struct extent_map *__lookup_extent_mapping(struct extent_map_tree *tree,
- u64 start, u64 len, int strict)
+static struct extent_map *
+__lookup_extent_mapping(struct extent_map_tree *tree,
+ u64 start, u64 len, int strict)
{
struct extent_map *em;
struct rb_node *rb_node;
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index c6598c89cff..61adc44b780 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -26,6 +26,7 @@ struct extent_map {
u64 mod_len;
u64 orig_start;
u64 orig_block_len;
+ u64 ram_bytes;
u64 block_start;
u64 block_len;
u64 generation;
@@ -61,7 +62,7 @@ void extent_map_tree_init(struct extent_map_tree *tree);
struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree,
u64 start, u64 len);
int add_extent_mapping(struct extent_map_tree *tree,
- struct extent_map *em);
+ struct extent_map *em, int modified);
int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em);
struct extent_map *alloc_extent_map(void);
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 94aa53b3872..b193bf324a4 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -83,10 +83,11 @@ out:
return ret;
}
-struct btrfs_csum_item *btrfs_lookup_csum(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
- u64 bytenr, int cow)
+static struct btrfs_csum_item *
+btrfs_lookup_csum(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct btrfs_path *path,
+ u64 bytenr, int cow)
{
int ret;
struct btrfs_key file_key;
@@ -118,9 +119,11 @@ struct btrfs_csum_item *btrfs_lookup_csum(struct btrfs_trans_handle *trans,
csums_in_item = btrfs_item_size_nr(leaf, path->slots[0]);
csums_in_item /= csum_size;
- if (csum_offset >= csums_in_item) {
+ if (csum_offset == csums_in_item) {
ret = -EFBIG;
goto fail;
+ } else if (csum_offset > csums_in_item) {
+ goto fail;
}
}
item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item);
@@ -150,32 +153,12 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
return ret;
}
-u64 btrfs_file_extent_length(struct btrfs_path *path)
-{
- int extent_type;
- struct btrfs_file_extent_item *fi;
- u64 len;
-
- fi = btrfs_item_ptr(path->nodes[0], path->slots[0],
- struct btrfs_file_extent_item);
- extent_type = btrfs_file_extent_type(path->nodes[0], fi);
-
- if (extent_type == BTRFS_FILE_EXTENT_REG ||
- extent_type == BTRFS_FILE_EXTENT_PREALLOC)
- len = btrfs_file_extent_num_bytes(path->nodes[0], fi);
- else if (extent_type == BTRFS_FILE_EXTENT_INLINE)
- len = btrfs_file_extent_inline_len(path->nodes[0], fi);
- else
- BUG();
-
- return len;
-}
-
static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
struct inode *inode, struct bio *bio,
u64 logical_offset, u32 *dst, int dio)
{
- u32 sum;
+ u32 sum[16];
+ int len;
struct bio_vec *bvec = bio->bi_io_vec;
int bio_index = 0;
u64 offset = 0;
@@ -184,7 +167,7 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
u64 disk_bytenr;
u32 diff;
u16 csum_size = btrfs_super_csum_size(root->fs_info->super_copy);
- int ret;
+ int count;
struct btrfs_path *path;
struct btrfs_csum_item *item = NULL;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
@@ -212,10 +195,12 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
if (dio)
offset = logical_offset;
while (bio_index < bio->bi_vcnt) {
+ len = min_t(int, ARRAY_SIZE(sum), bio->bi_vcnt - bio_index);
if (!dio)
offset = page_offset(bvec->bv_page) + bvec->bv_offset;
- ret = btrfs_find_ordered_sum(inode, offset, disk_bytenr, &sum);
- if (ret == 0)
+ count = btrfs_find_ordered_sum(inode, offset, disk_bytenr, sum,
+ len);
+ if (count)
goto found;
if (!item || disk_bytenr < item_start_offset ||
@@ -228,10 +213,8 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
item = btrfs_lookup_csum(NULL, root->fs_info->csum_root,
path, disk_bytenr, 0);
if (IS_ERR(item)) {
- ret = PTR_ERR(item);
- if (ret == -ENOENT || ret == -EFBIG)
- ret = 0;
- sum = 0;
+ count = 1;
+ sum[0] = 0;
if (BTRFS_I(inode)->root->root_key.objectid ==
BTRFS_DATA_RELOC_TREE_OBJECTID) {
set_extent_bits(io_tree, offset,
@@ -267,19 +250,29 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
diff = disk_bytenr - item_start_offset;
diff = diff / root->sectorsize;
diff = diff * csum_size;
-
- read_extent_buffer(path->nodes[0], &sum,
+ count = min_t(int, len, (item_last_offset - disk_bytenr) >>
+ inode->i_sb->s_blocksize_bits);
+ read_extent_buffer(path->nodes[0], sum,
((unsigned long)item) + diff,
- csum_size);
+ csum_size * count);
found:
- if (dst)
- *dst++ = sum;
- else
- set_state_private(io_tree, offset, sum);
- disk_bytenr += bvec->bv_len;
- offset += bvec->bv_len;
- bio_index++;
- bvec++;
+ if (dst) {
+ memcpy(dst, sum, count * csum_size);
+ dst += count;
+ } else {
+ if (dio)
+ extent_cache_csums_dio(io_tree, offset, sum,
+ count);
+ else
+ extent_cache_csums(io_tree, bio, bio_index, sum,
+ count);
+ }
+ while (count--) {
+ disk_bytenr += bvec->bv_len;
+ offset += bvec->bv_len;
+ bio_index++;
+ bvec++;
+ }
}
btrfs_free_path(path);
return 0;
@@ -356,11 +349,8 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
if (key.objectid != BTRFS_EXTENT_CSUM_OBJECTID ||
- key.type != BTRFS_EXTENT_CSUM_KEY)
- break;
-
- btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
- if (key.offset > end)
+ key.type != BTRFS_EXTENT_CSUM_KEY ||
+ key.offset > end)
break;
if (key.offset > start)
@@ -482,8 +472,7 @@ int btrfs_csum_one_bio(struct btrfs_root *root, struct inode *inode,
data = kmap_atomic(bvec->bv_page);
sector_sum->sum = ~(u32)0;
- sector_sum->sum = btrfs_csum_data(root,
- data + bvec->bv_offset,
+ sector_sum->sum = btrfs_csum_data(data + bvec->bv_offset,
sector_sum->sum,
bvec->bv_len);
kunmap_atomic(data);
@@ -516,8 +505,7 @@ int btrfs_csum_one_bio(struct btrfs_root *root, struct inode *inode,
* This calls btrfs_truncate_item with the correct args based on the
* overlap, and fixes up the key as required.
*/
-static noinline void truncate_one_csum(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+static noinline void truncate_one_csum(struct btrfs_root *root,
struct btrfs_path *path,
struct btrfs_key *key,
u64 bytenr, u64 len)
@@ -542,7 +530,7 @@ static noinline void truncate_one_csum(struct btrfs_trans_handle *trans,
*/
u32 new_size = (bytenr - key->offset) >> blocksize_bits;
new_size *= csum_size;
- btrfs_truncate_item(trans, root, path, new_size, 1);
+ btrfs_truncate_item(root, path, new_size, 1);
} else if (key->offset >= bytenr && csum_end > end_byte &&
end_byte > key->offset) {
/*
@@ -554,10 +542,10 @@ static noinline void truncate_one_csum(struct btrfs_trans_handle *trans,
u32 new_size = (csum_end - end_byte) >> blocksize_bits;
new_size *= csum_size;
- btrfs_truncate_item(trans, root, path, new_size, 0);
+ btrfs_truncate_item(root, path, new_size, 0);
key->offset = end_byte;
- btrfs_set_item_key_safe(trans, root, path, key);
+ btrfs_set_item_key_safe(root, path, key);
} else {
BUG();
}
@@ -672,7 +660,7 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
key.offset = end_byte - 1;
} else {
- truncate_one_csum(trans, root, path, &key, bytenr, len);
+ truncate_one_csum(root, path, &key, bytenr, len);
if (key.offset < bytenr)
break;
}
@@ -684,6 +672,24 @@ out:
return ret;
}
+static u64 btrfs_sector_sum_left(struct btrfs_ordered_sum *sums,
+ struct btrfs_sector_sum *sector_sum,
+ u64 total_bytes, u64 sectorsize)
+{
+ u64 tmp = sectorsize;
+ u64 next_sector = sector_sum->bytenr;
+ struct btrfs_sector_sum *next = sector_sum + 1;
+
+ while ((tmp + total_bytes) < sums->len) {
+ if (next_sector + sectorsize != next->bytenr)
+ break;
+ tmp += sectorsize;
+ next_sector = next->bytenr;
+ next++;
+ }
+ return tmp;
+}
+
int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_ordered_sum *sums)
@@ -710,7 +716,6 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
return -ENOMEM;
sector_sum = sums->sums;
- trans->adding_csums = 1;
again:
next_offset = (u64)-1;
found_next = 0;
@@ -789,22 +794,34 @@ again:
goto insert;
}
- if (csum_offset >= btrfs_item_size_nr(leaf, path->slots[0]) /
+ if (csum_offset == btrfs_item_size_nr(leaf, path->slots[0]) /
csum_size) {
- u32 diff = (csum_offset + 1) * csum_size;
+ int extend_nr;
+ u64 tmp;
+ u32 diff;
+ u32 free_space;
- /*
- * is the item big enough already? we dropped our lock
- * before and need to recheck
- */
- if (diff < btrfs_item_size_nr(leaf, path->slots[0]))
- goto csum;
+ if (btrfs_leaf_free_space(root, leaf) <
+ sizeof(struct btrfs_item) + csum_size * 2)
+ goto insert;
+
+ free_space = btrfs_leaf_free_space(root, leaf) -
+ sizeof(struct btrfs_item) - csum_size;
+ tmp = btrfs_sector_sum_left(sums, sector_sum, total_bytes,
+ root->sectorsize);
+ tmp >>= root->fs_info->sb->s_blocksize_bits;
+ WARN_ON(tmp < 1);
+
+ extend_nr = max_t(int, 1, (int)tmp);
+ diff = (csum_offset + extend_nr) * csum_size;
+ diff = min(diff, MAX_CSUM_ITEMS(root, csum_size) * csum_size);
diff = diff - btrfs_item_size_nr(leaf, path->slots[0]);
- if (diff != csum_size)
- goto insert;
+ diff = min(free_space, diff);
+ diff /= csum_size;
+ diff *= csum_size;
- btrfs_extend_item(trans, root, path, diff);
+ btrfs_extend_item(root, path, diff);
goto csum;
}
@@ -812,19 +829,14 @@ insert:
btrfs_release_path(path);
csum_offset = 0;
if (found_next) {
- u64 tmp = total_bytes + root->sectorsize;
- u64 next_sector = sector_sum->bytenr;
- struct btrfs_sector_sum *next = sector_sum + 1;
+ u64 tmp;
- while (tmp < sums->len) {
- if (next_sector + root->sectorsize != next->bytenr)
- break;
- tmp += root->sectorsize;
- next_sector = next->bytenr;
- next++;
- }
- tmp = min(tmp, next_offset - file_key.offset);
+ tmp = btrfs_sector_sum_left(sums, sector_sum, total_bytes,
+ root->sectorsize);
tmp >>= root->fs_info->sb->s_blocksize_bits;
+ tmp = min(tmp, (next_offset - file_key.offset) >>
+ root->fs_info->sb->s_blocksize_bits);
+
tmp = max((u64)1, tmp);
tmp = min(tmp, (u64)MAX_CSUM_ITEMS(root, csum_size));
ins_size = csum_size * tmp;
@@ -874,7 +886,6 @@ next_sector:
goto again;
}
out:
- trans->adding_csums = 0;
btrfs_free_path(path);
return ret;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index aeb84469d2c..4205ba752d4 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -24,17 +24,18 @@
#include <linux/string.h>
#include <linux/backing-dev.h>
#include <linux/mpage.h>
+#include <linux/aio.h>
#include <linux/falloc.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
#include <linux/slab.h>
+#include <linux/btrfs.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
-#include "ioctl.h"
#include "print-tree.h"
#include "tree-log.h"
#include "locking.h"
@@ -192,8 +193,8 @@ 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.
*/
-void btrfs_requeue_inode_defrag(struct inode *inode,
- struct inode_defrag *defrag)
+static void btrfs_requeue_inode_defrag(struct inode *inode,
+ struct inode_defrag *defrag)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
@@ -374,6 +375,11 @@ int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info)
atomic_inc(&fs_info->defrag_running);
while(1) {
+ /* Pause the auto defragger. */
+ if (test_bit(BTRFS_FS_STATE_REMOUNTING,
+ &fs_info->fs_state))
+ break;
+
if (!__need_auto_defrag(fs_info->tree_root))
break;
@@ -468,7 +474,7 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
/*
* unlocks pages after btrfs_file_write is done with them
*/
-void btrfs_drop_pages(struct page **pages, size_t num_pages)
+static void btrfs_drop_pages(struct page **pages, size_t num_pages)
{
size_t i;
for (i = 0; i < num_pages; i++) {
@@ -492,9 +498,9 @@ void btrfs_drop_pages(struct page **pages, size_t num_pages)
* doing real data extents, marking pages dirty and delalloc as required.
*/
int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
- struct page **pages, size_t num_pages,
- loff_t pos, size_t write_bytes,
- struct extent_state **cached)
+ struct page **pages, size_t num_pages,
+ loff_t pos, size_t write_bytes,
+ struct extent_state **cached)
{
int err = 0;
int i;
@@ -505,8 +511,7 @@ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
loff_t isize = i_size_read(inode);
start_pos = pos & ~((u64)root->sectorsize - 1);
- num_bytes = (write_bytes + pos - start_pos +
- root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
+ num_bytes = ALIGN(write_bytes + pos - start_pos, root->sectorsize);
end_of_last_block = start_pos + num_bytes - 1;
err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
@@ -548,6 +553,7 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
int testend = 1;
unsigned long flags;
int compressed = 0;
+ bool modified;
WARN_ON(end < start);
if (end == (u64)-1) {
@@ -557,6 +563,7 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
while (1) {
int no_splits = 0;
+ modified = false;
if (!split)
split = alloc_extent_map();
if (!split2)
@@ -587,6 +594,8 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
}
compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
+ clear_bit(EXTENT_FLAG_LOGGING, &flags);
+ modified = !list_empty(&em->list);
remove_extent_mapping(em_tree, em);
if (no_splits)
goto next;
@@ -602,15 +611,15 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
split->block_len = em->block_len;
else
split->block_len = split->len;
+ split->ram_bytes = em->ram_bytes;
split->orig_block_len = max(split->block_len,
em->orig_block_len);
split->generation = gen;
split->bdev = em->bdev;
split->flags = flags;
split->compress_type = em->compress_type;
- ret = add_extent_mapping(em_tree, split);
+ ret = add_extent_mapping(em_tree, split, modified);
BUG_ON(ret); /* Logic error */
- list_move(&split->list, &em_tree->modified_extents);
free_extent_map(split);
split = split2;
split2 = NULL;
@@ -627,6 +636,7 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
split->generation = gen;
split->orig_block_len = max(em->block_len,
em->orig_block_len);
+ split->ram_bytes = em->ram_bytes;
if (compressed) {
split->block_len = em->block_len;
@@ -638,9 +648,8 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
split->orig_start = em->orig_start;
}
- ret = add_extent_mapping(em_tree, split);
+ ret = add_extent_mapping(em_tree, split, modified);
BUG_ON(ret); /* Logic error */
- list_move(&split->list, &em_tree->modified_extents);
free_extent_map(split);
split = NULL;
}
@@ -816,7 +825,7 @@ next_slot:
memcpy(&new_key, &key, sizeof(new_key));
new_key.offset = end;
- btrfs_set_item_key_safe(trans, root, path, &new_key);
+ btrfs_set_item_key_safe(root, path, &new_key);
extent_offset += end - key.offset;
btrfs_set_file_extent_offset(leaf, fi, extent_offset);
@@ -1032,7 +1041,7 @@ again:
ino, bytenr, orig_offset,
&other_start, &other_end)) {
new_key.offset = end;
- btrfs_set_item_key_safe(trans, root, path, &new_key);
+ btrfs_set_item_key_safe(root, path, &new_key);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_generation(leaf, fi,
@@ -1066,7 +1075,7 @@ again:
trans->transid);
path->slots[0]++;
new_key.offset = start;
- btrfs_set_item_key_safe(trans, root, path, &new_key);
+ btrfs_set_item_key_safe(root, path, &new_key);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
@@ -1225,7 +1234,7 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file,
struct extent_state *cached_state = NULL;
int i;
unsigned long index = pos >> PAGE_CACHE_SHIFT;
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
int err = 0;
int faili = 0;
@@ -1312,7 +1321,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
struct iov_iter *i,
loff_t pos)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct page **pages = NULL;
unsigned long first_index;
@@ -1500,7 +1509,7 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
unsigned long nr_segs, loff_t pos)
{
struct file *file = iocb->ki_filp;
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
loff_t *ppos = &iocb->ki_pos;
u64 start_pos;
@@ -1509,8 +1518,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
size_t count, ocount;
bool sync = (file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host);
- sb_start_write(inode->i_sb);
-
mutex_lock(&inode->i_mutex);
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ);
@@ -1544,7 +1551,7 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
* although we have opened a file as writable, we have
* to stop this write operation to ensure FS consistency.
*/
- if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+ if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state)) {
mutex_unlock(&inode->i_mutex);
err = -EROFS;
goto out;
@@ -1612,7 +1619,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
if (sync)
atomic_dec(&BTRFS_I(inode)->sync_writers);
out:
- sb_end_write(inode->i_sb);
current->backing_dev_info = NULL;
return num_written ? num_written : err;
}
@@ -1627,7 +1633,20 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
*/
if (test_and_clear_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
&BTRFS_I(inode)->runtime_flags)) {
- btrfs_add_ordered_operation(NULL, BTRFS_I(inode)->root, inode);
+ struct btrfs_trans_handle *trans;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+
+ /*
+ * We need to block on a committing transaction to keep us from
+ * throwing a ordered operation on to the list and causing
+ * something like sync to deadlock trying to flush out this
+ * inode.
+ */
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans))
+ return PTR_ERR(trans);
+ btrfs_add_ordered_operation(trans, BTRFS_I(inode)->root, inode);
+ btrfs_end_transaction(trans, root);
if (inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
filemap_flush(inode->i_mapping);
}
@@ -1654,16 +1673,21 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret = 0;
struct btrfs_trans_handle *trans;
+ bool full_sync = 0;
trace_btrfs_sync_file(file, datasync);
/*
* We write the dirty pages in the range and wait until they complete
* out of the ->i_mutex. If so, we can flush the dirty pages by
- * multi-task, and make the performance up.
+ * multi-task, and make the performance up. See
+ * btrfs_wait_ordered_range for an explanation of the ASYNC check.
*/
atomic_inc(&BTRFS_I(inode)->sync_writers);
- ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
+ if (!ret && test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
+ &BTRFS_I(inode)->runtime_flags))
+ ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
atomic_dec(&BTRFS_I(inode)->sync_writers);
if (ret)
return ret;
@@ -1675,7 +1699,10 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
* range being left.
*/
atomic_inc(&root->log_batch);
- btrfs_wait_ordered_range(inode, start, end - start + 1);
+ full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags);
+ if (full_sync)
+ btrfs_wait_ordered_range(inode, start, end - start + 1);
atomic_inc(&root->log_batch);
/*
@@ -1742,13 +1769,25 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
if (ret != BTRFS_NO_LOG_SYNC) {
if (ret > 0) {
+ /*
+ * If we didn't already wait for ordered extents we need
+ * to do that now.
+ */
+ if (!full_sync)
+ btrfs_wait_ordered_range(inode, start,
+ end - start + 1);
ret = btrfs_commit_transaction(trans, root);
} else {
ret = btrfs_sync_log(trans, root);
- if (ret == 0)
+ if (ret == 0) {
ret = btrfs_end_transaction(trans, root);
- else
+ } else {
+ if (!full_sync)
+ btrfs_wait_ordered_range(inode, start,
+ end -
+ start + 1);
ret = btrfs_commit_transaction(trans, root);
+ }
}
} else {
ret = btrfs_end_transaction(trans, root);
@@ -1847,7 +1886,7 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
path->slots[0]++;
key.offset = offset;
- btrfs_set_item_key_safe(trans, root, path, &key);
+ btrfs_set_item_key_safe(root, path, &key);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
num_bytes = btrfs_file_extent_num_bytes(leaf, fi) + end -
@@ -1877,6 +1916,7 @@ out:
} else {
hole_em->start = offset;
hole_em->len = end - offset;
+ hole_em->ram_bytes = hole_em->len;
hole_em->orig_start = offset;
hole_em->block_start = EXTENT_MAP_HOLE;
@@ -1889,10 +1929,7 @@ out:
do {
btrfs_drop_extent_cache(inode, offset, end - 1, 0);
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, hole_em);
- if (!ret)
- list_move(&hole_em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, hole_em, 1);
write_unlock(&em_tree->lock);
} while (ret == -EEXIST);
free_extent_map(hole_em);
@@ -2102,8 +2139,9 @@ out:
static long btrfs_fallocate(struct file *file, int mode,
loff_t offset, loff_t len)
{
- struct inode *inode = file->f_path.dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct extent_state *cached_state = NULL;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
u64 cur_offset;
u64 last_byte;
u64 alloc_start;
@@ -2131,6 +2169,11 @@ static long btrfs_fallocate(struct file *file, int mode,
ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start);
if (ret)
return ret;
+ if (root->fs_info->quota_enabled) {
+ ret = btrfs_qgroup_reserve(root, alloc_end - alloc_start);
+ if (ret)
+ goto out_reserve_fail;
+ }
/*
* wait for ordered IO before we have any locks. We'll loop again
@@ -2234,6 +2277,9 @@ static long btrfs_fallocate(struct file *file, int mode,
&cached_state, GFP_NOFS);
out:
mutex_unlock(&inode->i_mutex);
+ if (root->fs_info->quota_enabled)
+ btrfs_qgroup_free(root, alloc_end - alloc_start);
+out_reserve_fail:
/* Let go of our reservation. */
btrfs_free_reserved_data_space(inode, alloc_end - alloc_start);
return ret;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 0be7a8742a4..ecca6c7375a 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -104,7 +104,8 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
spin_lock(&block_group->lock);
if (!((BTRFS_I(inode)->flags & flags) == flags)) {
- printk(KERN_INFO "Old style space inode found, converting.\n");
+ btrfs_info(root->fs_info,
+ "Old style space inode found, converting.");
BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM |
BTRFS_INODE_NODATACOW;
block_group->disk_cache_state = BTRFS_DC_CLEAR;
@@ -119,9 +120,10 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
return inode;
}
-int __create_free_space_inode(struct btrfs_root *root,
- struct btrfs_trans_handle *trans,
- struct btrfs_path *path, u64 ino, u64 offset)
+static int __create_free_space_inode(struct btrfs_root *root,
+ struct btrfs_trans_handle *trans,
+ struct btrfs_path *path,
+ u64 ino, u64 offset)
{
struct btrfs_key key;
struct btrfs_disk_key disk_key;
@@ -431,7 +433,7 @@ static void io_ctl_set_crc(struct io_ctl *io_ctl, int index)
if (index == 0)
offset = sizeof(u32) * io_ctl->num_pages;
- crc = btrfs_csum_data(io_ctl->root, io_ctl->orig + offset, crc,
+ crc = btrfs_csum_data(io_ctl->orig + offset, crc,
PAGE_CACHE_SIZE - offset);
btrfs_csum_final(crc, (char *)&crc);
io_ctl_unmap_page(io_ctl);
@@ -461,7 +463,7 @@ static int io_ctl_check_crc(struct io_ctl *io_ctl, int index)
kunmap(io_ctl->pages[0]);
io_ctl_map_page(io_ctl, 0);
- crc = btrfs_csum_data(io_ctl->root, io_ctl->orig + offset, crc,
+ crc = btrfs_csum_data(io_ctl->orig + offset, crc,
PAGE_CACHE_SIZE - offset);
btrfs_csum_final(crc, (char *)&crc);
if (val != crc) {
@@ -624,9 +626,9 @@ next:
spin_unlock(&ctl->tree_lock);
}
-int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
- struct btrfs_free_space_ctl *ctl,
- struct btrfs_path *path, u64 offset)
+static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
+ struct btrfs_free_space_ctl *ctl,
+ struct btrfs_path *path, u64 offset)
{
struct btrfs_free_space_header *header;
struct extent_buffer *leaf;
@@ -669,10 +671,11 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
btrfs_release_path(path);
if (BTRFS_I(inode)->generation != generation) {
- printk(KERN_ERR "btrfs: free space inode generation (%llu) did"
- " not match free space cache generation (%llu)\n",
- (unsigned long long)BTRFS_I(inode)->generation,
- (unsigned long long)generation);
+ btrfs_err(root->fs_info,
+ "free space inode generation (%llu) "
+ "did not match free space cache generation (%llu)",
+ (unsigned long long)BTRFS_I(inode)->generation,
+ (unsigned long long)generation);
return 0;
}
@@ -721,8 +724,8 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
ret = link_free_space(ctl, e);
spin_unlock(&ctl->tree_lock);
if (ret) {
- printk(KERN_ERR "Duplicate entries in "
- "free space cache, dumping\n");
+ btrfs_err(root->fs_info,
+ "Duplicate entries in free space cache, dumping");
kmem_cache_free(btrfs_free_space_cachep, e);
goto free_cache;
}
@@ -741,8 +744,8 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
ctl->op->recalc_thresholds(ctl);
spin_unlock(&ctl->tree_lock);
if (ret) {
- printk(KERN_ERR "Duplicate entries in "
- "free space cache, dumping\n");
+ btrfs_err(root->fs_info,
+ "Duplicate entries in free space cache, dumping");
kmem_cache_free(btrfs_free_space_cachep, e);
goto free_cache;
}
@@ -833,8 +836,8 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
if (!matched) {
__btrfs_remove_free_space_cache(ctl);
- printk(KERN_ERR "block group %llu has an wrong amount of free "
- "space\n", block_group->key.objectid);
+ btrfs_err(fs_info, "block group %llu has wrong amount of free space",
+ block_group->key.objectid);
ret = -1;
}
out:
@@ -845,8 +848,8 @@ out:
spin_unlock(&block_group->lock);
ret = 0;
- printk(KERN_ERR "btrfs: failed to load free space cache "
- "for block group %llu\n", block_group->key.objectid);
+ btrfs_err(fs_info, "failed to load free space cache for block group %llu",
+ block_group->key.objectid);
}
iput(inode);
@@ -866,11 +869,11 @@ out:
* on mount. This will return 0 if it was successfull in writing the cache out,
* and -1 if it was not.
*/
-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_trans_handle *trans,
- struct btrfs_path *path, u64 offset)
+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_trans_handle *trans,
+ struct btrfs_path *path, u64 offset)
{
struct btrfs_free_space_header *header;
struct extent_buffer *leaf;
@@ -1104,8 +1107,9 @@ int btrfs_write_out_cache(struct btrfs_root *root,
spin_unlock(&block_group->lock);
ret = 0;
#ifdef DEBUG
- printk(KERN_ERR "btrfs: failed to write free space cache "
- "for block group %llu\n", block_group->key.objectid);
+ btrfs_err(root->fs_info,
+ "failed to write free space cache for block group %llu",
+ block_group->key.objectid);
#endif
}
@@ -1356,6 +1360,8 @@ static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl)
u64 bytes_per_bg = BITS_PER_BITMAP * ctl->unit;
int max_bitmaps = div64_u64(size + bytes_per_bg - 1, bytes_per_bg);
+ max_bitmaps = max(max_bitmaps, 1);
+
BUG_ON(ctl->total_bitmaps > max_bitmaps);
/*
@@ -1463,10 +1469,14 @@ static int search_bitmap(struct btrfs_free_space_ctl *ctl,
}
static struct btrfs_free_space *
-find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes)
+find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
+ unsigned long align)
{
struct btrfs_free_space *entry;
struct rb_node *node;
+ u64 ctl_off;
+ u64 tmp;
+ u64 align_off;
int ret;
if (!ctl->free_space_offset.rb_node)
@@ -1481,15 +1491,34 @@ find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes)
if (entry->bytes < *bytes)
continue;
+ /* make sure the space returned is big enough
+ * to match our requested alignment
+ */
+ if (*bytes >= align) {
+ ctl_off = entry->offset - ctl->start;
+ tmp = ctl_off + align - 1;;
+ do_div(tmp, align);
+ tmp = tmp * align + ctl->start;
+ align_off = tmp - entry->offset;
+ } else {
+ align_off = 0;
+ tmp = entry->offset;
+ }
+
+ if (entry->bytes < *bytes + align_off)
+ continue;
+
if (entry->bitmap) {
- ret = search_bitmap(ctl, entry, offset, bytes);
- if (!ret)
+ ret = search_bitmap(ctl, entry, &tmp, bytes);
+ if (!ret) {
+ *offset = tmp;
return entry;
+ }
continue;
}
- *offset = entry->offset;
- *bytes = entry->bytes;
+ *offset = tmp;
+ *bytes = entry->bytes - align_off;
return entry;
}
@@ -1539,7 +1568,8 @@ again:
search_bytes = ctl->unit;
search_bytes = min(search_bytes, end - search_start + 1);
ret = search_bitmap(ctl, bitmap_info, &search_start, &search_bytes);
- BUG_ON(ret < 0 || search_start != *offset);
+ if (ret < 0 || search_start != *offset)
+ return -EINVAL;
/* We may have found more bits than what we need */
search_bytes = min(search_bytes, *bytes);
@@ -1636,10 +1666,14 @@ static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
}
/*
- * some block groups are so tiny they can't be enveloped by a bitmap, so
- * don't even bother to create a bitmap for this
+ * The original block groups from mkfs can be really small, like 8
+ * megabytes, so don't bother with a bitmap for those entries. However
+ * some block groups can be smaller than what a bitmap would cover but
+ * are still large enough that they could overflow the 32k memory limit,
+ * so allow those block groups to still be allowed to have a bitmap
+ * entry.
*/
- if (BITS_PER_BITMAP * ctl->unit > block_group->key.offset)
+ if (((BITS_PER_BITMAP * ctl->unit) >> 1) > block_group->key.offset)
return false;
return true;
@@ -1941,7 +1975,6 @@ again:
re_search = true;
goto again;
}
- BUG_ON(ret); /* logic error */
out_lock:
spin_unlock(&ctl->tree_lock);
out:
@@ -2035,7 +2068,8 @@ out:
return 0;
}
-void __btrfs_remove_free_space_cache_locked(struct btrfs_free_space_ctl *ctl)
+static void __btrfs_remove_free_space_cache_locked(
+ struct btrfs_free_space_ctl *ctl)
{
struct btrfs_free_space *info;
struct rb_node *node;
@@ -2095,9 +2129,12 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
struct btrfs_free_space *entry = NULL;
u64 bytes_search = bytes + empty_size;
u64 ret = 0;
+ u64 align_gap = 0;
+ u64 align_gap_len = 0;
spin_lock(&ctl->tree_lock);
- entry = find_free_space(ctl, &offset, &bytes_search);
+ entry = find_free_space(ctl, &offset, &bytes_search,
+ block_group->full_stripe_len);
if (!entry)
goto out;
@@ -2107,9 +2144,15 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
if (!entry->bytes)
free_bitmap(ctl, entry);
} else {
+
unlink_free_space(ctl, entry);
- entry->offset += bytes;
- entry->bytes -= bytes;
+ align_gap_len = offset - entry->offset;
+ align_gap = entry->offset;
+
+ entry->offset = offset + bytes;
+ WARN_ON(entry->bytes < bytes + align_gap_len);
+
+ entry->bytes -= bytes + align_gap_len;
if (!entry->bytes)
kmem_cache_free(btrfs_free_space_cachep, entry);
else
@@ -2119,6 +2162,8 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
out:
spin_unlock(&ctl->tree_lock);
+ if (align_gap_len)
+ __btrfs_add_free_space(ctl, align_gap, align_gap_len);
return ret;
}
@@ -2891,8 +2936,9 @@ int load_free_ino_cache(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
ret = __load_free_space_cache(root, inode, ctl, path, 0);
if (ret < 0)
- printk(KERN_ERR "btrfs: failed to load free ino cache for "
- "root %llu\n", root->root_key.objectid);
+ btrfs_err(fs_info,
+ "failed to load free ino cache for root %llu",
+ root->root_key.objectid);
out_put:
iput(inode);
out:
@@ -2919,11 +2965,531 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
if (ret) {
btrfs_delalloc_release_metadata(inode, inode->i_size);
#ifdef DEBUG
- printk(KERN_ERR "btrfs: failed to write free ino cache "
- "for root %llu\n", root->root_key.objectid);
+ btrfs_err(root->fs_info,
+ "failed to write free ino cache for root %llu",
+ root->root_key.objectid);
#endif
}
iput(inode);
return ret;
}
+
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+static struct btrfs_block_group_cache *init_test_block_group(void)
+{
+ struct btrfs_block_group_cache *cache;
+
+ cache = kzalloc(sizeof(*cache), GFP_NOFS);
+ if (!cache)
+ return NULL;
+ cache->free_space_ctl = kzalloc(sizeof(*cache->free_space_ctl),
+ GFP_NOFS);
+ if (!cache->free_space_ctl) {
+ kfree(cache);
+ return NULL;
+ }
+
+ cache->key.objectid = 0;
+ cache->key.offset = 1024 * 1024 * 1024;
+ cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+ cache->sectorsize = 4096;
+
+ spin_lock_init(&cache->lock);
+ INIT_LIST_HEAD(&cache->list);
+ INIT_LIST_HEAD(&cache->cluster_list);
+ INIT_LIST_HEAD(&cache->new_bg_list);
+
+ btrfs_init_free_space_ctl(cache);
+
+ return cache;
+}
+
+/*
+ * Checks to see if the given range is in the free space cache. This is really
+ * just used to check the absence of space, so if there is free space in the
+ * range at all we will return 1.
+ */
+static int check_exists(struct btrfs_block_group_cache *cache, u64 offset,
+ u64 bytes)
+{
+ struct btrfs_free_space_ctl *ctl = cache->free_space_ctl;
+ struct btrfs_free_space *info;
+ int ret = 0;
+
+ spin_lock(&ctl->tree_lock);
+ info = tree_search_offset(ctl, offset, 0, 0);
+ if (!info) {
+ info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset),
+ 1, 0);
+ if (!info)
+ goto out;
+ }
+
+have_info:
+ if (info->bitmap) {
+ u64 bit_off, bit_bytes;
+ struct rb_node *n;
+ struct btrfs_free_space *tmp;
+
+ bit_off = offset;
+ bit_bytes = ctl->unit;
+ ret = search_bitmap(ctl, info, &bit_off, &bit_bytes);
+ if (!ret) {
+ if (bit_off == offset) {
+ ret = 1;
+ goto out;
+ } else if (bit_off > offset &&
+ offset + bytes > bit_off) {
+ ret = 1;
+ goto out;
+ }
+ }
+
+ n = rb_prev(&info->offset_index);
+ while (n) {
+ tmp = rb_entry(n, struct btrfs_free_space,
+ offset_index);
+ if (tmp->offset + tmp->bytes < offset)
+ break;
+ if (offset + bytes < tmp->offset) {
+ n = rb_prev(&info->offset_index);
+ continue;
+ }
+ info = tmp;
+ goto have_info;
+ }
+
+ n = rb_next(&info->offset_index);
+ while (n) {
+ tmp = rb_entry(n, struct btrfs_free_space,
+ offset_index);
+ if (offset + bytes < tmp->offset)
+ break;
+ if (tmp->offset + tmp->bytes < offset) {
+ n = rb_next(&info->offset_index);
+ continue;
+ }
+ info = tmp;
+ goto have_info;
+ }
+
+ goto out;
+ }
+
+ if (info->offset == offset) {
+ ret = 1;
+ goto out;
+ }
+
+ if (offset > info->offset && offset < info->offset + info->bytes)
+ ret = 1;
+out:
+ spin_unlock(&ctl->tree_lock);
+ return ret;
+}
+
+/*
+ * Use this if you need to make a bitmap or extent entry specifically, it
+ * doesn't do any of the merging that add_free_space does, this acts a lot like
+ * how the free space cache loading stuff works, so you can get really weird
+ * configurations.
+ */
+static int add_free_space_entry(struct btrfs_block_group_cache *cache,
+ u64 offset, u64 bytes, bool bitmap)
+{
+ struct btrfs_free_space_ctl *ctl = cache->free_space_ctl;
+ struct btrfs_free_space *info = NULL, *bitmap_info;
+ void *map = NULL;
+ u64 bytes_added;
+ int ret;
+
+again:
+ if (!info) {
+ info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS);
+ if (!info)
+ return -ENOMEM;
+ }
+
+ if (!bitmap) {
+ spin_lock(&ctl->tree_lock);
+ info->offset = offset;
+ info->bytes = bytes;
+ ret = link_free_space(ctl, info);
+ spin_unlock(&ctl->tree_lock);
+ if (ret)
+ kmem_cache_free(btrfs_free_space_cachep, info);
+ return ret;
+ }
+
+ if (!map) {
+ map = kzalloc(PAGE_CACHE_SIZE, GFP_NOFS);
+ if (!map) {
+ kmem_cache_free(btrfs_free_space_cachep, info);
+ return -ENOMEM;
+ }
+ }
+
+ spin_lock(&ctl->tree_lock);
+ bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset),
+ 1, 0);
+ if (!bitmap_info) {
+ info->bitmap = map;
+ map = NULL;
+ add_new_bitmap(ctl, info, offset);
+ bitmap_info = info;
+ }
+
+ bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes);
+ bytes -= bytes_added;
+ offset += bytes_added;
+ spin_unlock(&ctl->tree_lock);
+
+ if (bytes)
+ goto again;
+
+ if (map)
+ kfree(map);
+ return 0;
+}
+
+/*
+ * This test just does basic sanity checking, making sure we can add an exten
+ * entry and remove space from either end and the middle, and make sure we can
+ * remove space that covers adjacent extent entries.
+ */
+static int test_extents(struct btrfs_block_group_cache *cache)
+{
+ int ret = 0;
+
+ printk(KERN_ERR "Running extent only tests\n");
+
+ /* First just make sure we can remove an entire entry */
+ ret = btrfs_add_free_space(cache, 0, 4 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error adding initial extents %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 0, 4 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error removing extent %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 0, 4 * 1024 * 1024)) {
+ printk(KERN_ERR "Full remove left some lingering space\n");
+ return -1;
+ }
+
+ /* Ok edge and middle cases now */
+ ret = btrfs_add_free_space(cache, 0, 4 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error adding half extent %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 3 * 1024 * 1024, 1 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error removing tail end %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 0, 1 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error removing front end %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 2 * 1024 * 1024, 4096);
+ if (ret) {
+ printk(KERN_ERR "Error removing middle peice %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 0, 1 * 1024 * 1024)) {
+ printk(KERN_ERR "Still have space at the front\n");
+ return -1;
+ }
+
+ if (check_exists(cache, 2 * 1024 * 1024, 4096)) {
+ printk(KERN_ERR "Still have space in the middle\n");
+ return -1;
+ }
+
+ if (check_exists(cache, 3 * 1024 * 1024, 1 * 1024 * 1024)) {
+ printk(KERN_ERR "Still have space at the end\n");
+ return -1;
+ }
+
+ /* Cleanup */
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+
+ return 0;
+}
+
+static int test_bitmaps(struct btrfs_block_group_cache *cache)
+{
+ u64 next_bitmap_offset;
+ int ret;
+
+ printk(KERN_ERR "Running bitmap only tests\n");
+
+ ret = add_free_space_entry(cache, 0, 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't create a bitmap entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 0, 4 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error removing bitmap full range %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 0, 4 * 1024 * 1024)) {
+ printk(KERN_ERR "Left some space in bitmap\n");
+ return -1;
+ }
+
+ ret = add_free_space_entry(cache, 0, 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add to our bitmap entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 1 * 1024 * 1024, 2 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Couldn't remove middle chunk %d\n", ret);
+ return ret;
+ }
+
+ /*
+ * The first bitmap we have starts at offset 0 so the next one is just
+ * at the end of the first bitmap.
+ */
+ next_bitmap_offset = (u64)(BITS_PER_BITMAP * 4096);
+
+ /* Test a bit straddling two bitmaps */
+ ret = add_free_space_entry(cache, next_bitmap_offset -
+ (2 * 1024 * 1024), 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add space that straddles two bitmaps"
+ " %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, next_bitmap_offset -
+ (1 * 1024 * 1024), 2 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Couldn't remove overlapping space %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, next_bitmap_offset - (1 * 1024 * 1024),
+ 2 * 1024 * 1024)) {
+ printk(KERN_ERR "Left some space when removing overlapping\n");
+ return -1;
+ }
+
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+
+ return 0;
+}
+
+/* This is the high grade jackassery */
+static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache)
+{
+ u64 bitmap_offset = (u64)(BITS_PER_BITMAP * 4096);
+ int ret;
+
+ printk(KERN_ERR "Running bitmap and extent tests\n");
+
+ /*
+ * First let's do something simple, an extent at the same offset as the
+ * bitmap, but the free space completely in the extent and then
+ * completely in the bitmap.
+ */
+ ret = add_free_space_entry(cache, 4 * 1024 * 1024, 1 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't create bitmap entry %d\n", ret);
+ return ret;
+ }
+
+ ret = add_free_space_entry(cache, 0, 1 * 1024 * 1024, 0);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add extent entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 0, 1 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Couldn't remove extent entry %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 0, 1 * 1024 * 1024)) {
+ printk(KERN_ERR "Left remnants after our remove\n");
+ return -1;
+ }
+
+ /* Now to add back the extent entry and remove from the bitmap */
+ ret = add_free_space_entry(cache, 0, 1 * 1024 * 1024, 0);
+ if (ret) {
+ printk(KERN_ERR "Couldn't re-add extent entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 4 * 1024 * 1024, 1 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Couldn't remove from bitmap %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 4 * 1024 * 1024, 1 * 1024 * 1024)) {
+ printk(KERN_ERR "Left remnants in the bitmap\n");
+ return -1;
+ }
+
+ /*
+ * Ok so a little more evil, extent entry and bitmap at the same offset,
+ * removing an overlapping chunk.
+ */
+ ret = add_free_space_entry(cache, 1 * 1024 * 1024, 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add to a bitmap %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 512 * 1024, 3 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Couldn't remove overlapping space %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 512 * 1024, 3 * 1024 * 1024)) {
+ printk(KERN_ERR "Left over peices after removing "
+ "overlapping\n");
+ return -1;
+ }
+
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+
+ /* Now with the extent entry offset into the bitmap */
+ ret = add_free_space_entry(cache, 4 * 1024 * 1024, 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add space to the bitmap %d\n", ret);
+ return ret;
+ }
+
+ ret = add_free_space_entry(cache, 2 * 1024 * 1024, 2 * 1024 * 1024, 0);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add extent to the cache %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 3 * 1024 * 1024, 4 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Problem removing overlapping space %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, 3 * 1024 * 1024, 4 * 1024 * 1024)) {
+ printk(KERN_ERR "Left something behind when removing space");
+ return -1;
+ }
+
+ /*
+ * This has blown up in the past, the extent entry starts before the
+ * bitmap entry, but we're trying to remove an offset that falls
+ * completely within the bitmap range and is in both the extent entry
+ * and the bitmap entry, looks like this
+ *
+ * [ extent ]
+ * [ bitmap ]
+ * [ del ]
+ */
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+ ret = add_free_space_entry(cache, bitmap_offset + 4 * 1024 * 1024,
+ 4 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add bitmap %d\n", ret);
+ return ret;
+ }
+
+ ret = add_free_space_entry(cache, bitmap_offset - 1 * 1024 * 1024,
+ 5 * 1024 * 1024, 0);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add extent entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, bitmap_offset + 1 * 1024 * 1024,
+ 5 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Failed to free our space %d\n", ret);
+ return ret;
+ }
+
+ if (check_exists(cache, bitmap_offset + 1 * 1024 * 1024,
+ 5 * 1024 * 1024)) {
+ printk(KERN_ERR "Left stuff over\n");
+ return -1;
+ }
+
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+
+ /*
+ * This blew up before, we have part of the free space in a bitmap and
+ * then the entirety of the rest of the space in an extent. This used
+ * to return -EAGAIN back from btrfs_remove_extent, make sure this
+ * doesn't happen.
+ */
+ ret = add_free_space_entry(cache, 1 * 1024 * 1024, 2 * 1024 * 1024, 1);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add bitmap entry %d\n", ret);
+ return ret;
+ }
+
+ ret = add_free_space_entry(cache, 3 * 1024 * 1024, 1 * 1024 * 1024, 0);
+ if (ret) {
+ printk(KERN_ERR "Couldn't add extent entry %d\n", ret);
+ return ret;
+ }
+
+ ret = btrfs_remove_free_space(cache, 1 * 1024 * 1024, 3 * 1024 * 1024);
+ if (ret) {
+ printk(KERN_ERR "Error removing bitmap and extent "
+ "overlapping %d\n", ret);
+ return ret;
+ }
+
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+ return 0;
+}
+
+void btrfs_test_free_space_cache(void)
+{
+ struct btrfs_block_group_cache *cache;
+
+ printk(KERN_ERR "Running btrfs free space cache tests\n");
+
+ cache = init_test_block_group();
+ if (!cache) {
+ printk(KERN_ERR "Couldn't run the tests\n");
+ return;
+ }
+
+ if (test_extents(cache))
+ goto out;
+ if (test_bitmaps(cache))
+ goto out;
+ if (test_bitmaps_and_extents(cache))
+ goto out;
+out:
+ __btrfs_remove_free_space_cache(cache->free_space_ctl);
+ kfree(cache->free_space_ctl);
+ kfree(cache);
+ printk(KERN_ERR "Free space cache tests finished\n");
+}
+#endif /* CONFIG_BTRFS_FS_RUN_SANITY_TESTS */
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index 8f2613f779e..4dc17d8809c 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -110,4 +110,9 @@ int btrfs_return_cluster_to_free_space(
struct btrfs_free_cluster *cluster);
int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group,
u64 *trimmed, u64 start, u64 end, u64 minlen);
+
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+void btrfs_test_free_space_cache(void);
+#endif
+
#endif
diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c
index 48b8fda9313..e0b7034d634 100644
--- a/fs/btrfs/inode-item.c
+++ b/fs/btrfs/inode-item.c
@@ -183,10 +183,11 @@ int btrfs_get_inode_ref_index(struct btrfs_trans_handle *trans,
return -ENOENT;
}
-int btrfs_del_inode_extref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- const char *name, int name_len,
- u64 inode_objectid, u64 ref_objectid, u64 *index)
+static int btrfs_del_inode_extref(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ const char *name, int name_len,
+ u64 inode_objectid, u64 ref_objectid,
+ u64 *index)
{
struct btrfs_path *path;
struct btrfs_key key;
@@ -246,7 +247,7 @@ int btrfs_del_inode_extref(struct btrfs_trans_handle *trans,
memmove_extent_buffer(leaf, ptr, ptr + del_len,
item_size - (ptr + del_len - item_start));
- btrfs_truncate_item(trans, root, path, item_size - del_len, 1);
+ btrfs_truncate_item(root, path, item_size - del_len, 1);
out:
btrfs_free_path(path);
@@ -309,7 +310,7 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans,
item_start = btrfs_item_ptr_offset(leaf, path->slots[0]);
memmove_extent_buffer(leaf, ptr, ptr + sub_item_len,
item_size - (ptr + sub_item_len - item_start));
- btrfs_truncate_item(trans, root, path, item_size - sub_item_len, 1);
+ btrfs_truncate_item(root, path, item_size - sub_item_len, 1);
out:
btrfs_free_path(path);
@@ -361,7 +362,7 @@ static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans,
name, name_len, NULL))
goto out;
- btrfs_extend_item(trans, root, path, ins_len);
+ btrfs_extend_item(root, path, ins_len);
ret = 0;
}
if (ret < 0)
@@ -417,7 +418,7 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans,
goto out;
old_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
- btrfs_extend_item(trans, root, path, ins_len);
+ btrfs_extend_item(root, path, ins_len);
ref = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_ref);
ref = (struct btrfs_inode_ref *)((unsigned long)ref + old_size);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index cc93b23ca35..9b31b3b091f 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -32,6 +32,7 @@
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
+#include <linux/aio.h>
#include <linux/bit_spinlock.h>
#include <linux/xattr.h>
#include <linux/posix_acl.h>
@@ -39,12 +40,13 @@
#include <linux/slab.h>
#include <linux/ratelimit.h>
#include <linux/mount.h>
+#include <linux/btrfs.h>
+#include <linux/blkdev.h>
#include "compat.h"
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
-#include "ioctl.h"
#include "print-tree.h"
#include "ordered-data.h"
#include "xattr.h"
@@ -54,6 +56,7 @@
#include "locking.h"
#include "free-space-cache.h"
#include "inode-map.h"
+#include "backref.h"
struct btrfs_iget_args {
u64 ino;
@@ -98,7 +101,10 @@ static noinline int cow_file_range(struct inode *inode,
static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
u64 len, u64 orig_start,
u64 block_start, u64 block_len,
- u64 orig_block_len, int type);
+ u64 orig_block_len, u64 ram_bytes,
+ int type);
+
+static int btrfs_dirty_inode(struct inode *inode);
static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
struct inode *inode, struct inode *dir,
@@ -231,8 +237,7 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans,
u64 isize = i_size_read(inode);
u64 actual_end = min(end + 1, isize);
u64 inline_len = actual_end - start;
- u64 aligned_end = (end + root->sectorsize - 1) &
- ~((u64)root->sectorsize - 1);
+ u64 aligned_end = ALIGN(end, root->sectorsize);
u64 data_len = inline_len;
int ret;
@@ -265,6 +270,7 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans,
return 1;
}
+ 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);
return 0;
@@ -351,6 +357,7 @@ static noinline int compress_file_range(struct inode *inode,
int i;
int will_compress;
int compress_type = root->fs_info->compress_type;
+ int redirty = 0;
/* if this is a small write inside eof, kick off a defrag */
if ((end - start + 1) < 16 * 1024 &&
@@ -389,7 +396,7 @@ again:
* a compressed extent to 128k.
*/
total_compressed = min(total_compressed, max_uncompressed);
- num_bytes = (end - start + blocksize) & ~(blocksize - 1);
+ num_bytes = ALIGN(end - start + 1, blocksize);
num_bytes = max(blocksize, num_bytes);
total_in = 0;
ret = 0;
@@ -413,6 +420,17 @@ again:
if (BTRFS_I(inode)->force_compress)
compress_type = BTRFS_I(inode)->force_compress;
+ /*
+ * we need to call clear_page_dirty_for_io on each
+ * page in the range. Otherwise applications with the file
+ * mmap'd can wander in and change the page contents while
+ * we are compressing them.
+ *
+ * If the compression fails for any reason, we set the pages
+ * dirty again later on.
+ */
+ extent_range_clear_dirty_for_io(inode, start, end);
+ redirty = 1;
ret = btrfs_compress_pages(compress_type,
inode->i_mapping, start,
total_compressed, pages,
@@ -488,15 +506,13 @@ cont:
* up to a block size boundary so the allocator does sane
* things
*/
- total_compressed = (total_compressed + blocksize - 1) &
- ~(blocksize - 1);
+ total_compressed = ALIGN(total_compressed, blocksize);
/*
* one last check to make sure the compression is really a
* win, compare the page count read with the blocks on disk
*/
- total_in = (total_in + PAGE_CACHE_SIZE - 1) &
- ~(PAGE_CACHE_SIZE - 1);
+ total_in = ALIGN(total_in, PAGE_CACHE_SIZE);
if (total_compressed >= total_in) {
will_compress = 0;
} else {
@@ -554,6 +570,8 @@ cleanup_and_bail_uncompressed:
__set_page_dirty_nobuffers(locked_page);
/* unlocked later on in the async handlers */
}
+ if (redirty)
+ extent_range_redirty_for_io(inode, start, end);
add_async_extent(async_cow, start, end - start + 1,
0, NULL, 0, BTRFS_COMPRESS_NONE);
*num_added += 1;
@@ -608,7 +626,7 @@ static noinline int submit_compressed_extents(struct inode *inode,
if (list_empty(&async_cow->extents))
return 0;
-
+again:
while (!list_empty(&async_cow->extents)) {
async_extent = list_entry(async_cow->extents.next,
struct async_extent, list);
@@ -648,6 +666,8 @@ retry:
async_extent->ram_size - 1,
btrfs_get_extent,
WB_SYNC_ALL);
+ else if (ret)
+ unlock_page(async_cow->locked_page);
kfree(async_extent);
cond_resched();
continue;
@@ -672,6 +692,7 @@ retry:
if (ret) {
int i;
+
for (i = 0; i < async_extent->nr_pages; i++) {
WARN_ON(async_extent->pages[i]->mapping);
page_cache_release(async_extent->pages[i]);
@@ -679,12 +700,10 @@ retry:
kfree(async_extent->pages);
async_extent->nr_pages = 0;
async_extent->pages = NULL;
- unlock_extent(io_tree, async_extent->start,
- async_extent->start +
- async_extent->ram_size - 1);
+
if (ret == -ENOSPC)
goto retry;
- goto out_free; /* JDM: Requeue? */
+ goto out_free;
}
/*
@@ -696,14 +715,18 @@ retry:
async_extent->ram_size - 1, 0);
em = alloc_extent_map();
- BUG_ON(!em); /* -ENOMEM */
+ if (!em)
+ 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 = root->fs_info->fs_devices->latest_bdev;
em->compress_type = async_extent->compress_type;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
@@ -712,10 +735,7 @@ retry:
while (1) {
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- if (!ret)
- list_move(&em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
@@ -726,6 +746,9 @@ retry:
async_extent->ram_size - 1, 0);
}
+ if (ret)
+ goto out_free_reserve;
+
ret = btrfs_add_ordered_extent_compress(inode,
async_extent->start,
ins.objectid,
@@ -733,7 +756,8 @@ retry:
ins.offset,
BTRFS_ORDERED_COMPRESSED,
async_extent->compress_type);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ goto out_free_reserve;
/*
* clear dirty, set writeback and unlock the pages.
@@ -754,18 +778,30 @@ retry:
ins.objectid,
ins.offset, async_extent->pages,
async_extent->nr_pages);
-
- BUG_ON(ret); /* -ENOMEM */
alloc_hint = ins.objectid + ins.offset;
kfree(async_extent);
+ if (ret)
+ goto out;
cond_resched();
}
ret = 0;
out:
return ret;
+out_free_reserve:
+ btrfs_free_reserved_extent(root, ins.objectid, ins.offset);
out_free:
+ extent_clear_unlock_delalloc(inode, &BTRFS_I(inode)->io_tree,
+ async_extent->start,
+ async_extent->start +
+ async_extent->ram_size - 1,
+ NULL, EXTENT_CLEAR_UNLOCK_PAGE |
+ EXTENT_CLEAR_UNLOCK |
+ EXTENT_CLEAR_DELALLOC |
+ EXTENT_CLEAR_DIRTY |
+ EXTENT_SET_WRITEBACK |
+ EXTENT_END_WRITEBACK);
kfree(async_extent);
- goto out;
+ goto again;
}
static u64 get_extent_allocation_hint(struct inode *inode, u64 start,
@@ -834,7 +870,7 @@ static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
BUG_ON(btrfs_is_free_space_inode(inode));
- num_bytes = (end - start + blocksize) & ~(blocksize - 1);
+ num_bytes = ALIGN(end - start + 1, blocksize);
num_bytes = max(blocksize, num_bytes);
disk_num_bytes = num_bytes;
@@ -887,25 +923,26 @@ static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
}
em = alloc_extent_map();
- BUG_ON(!em); /* -ENOMEM */
+ if (!em)
+ 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 = root->fs_info->fs_devices->latest_bdev;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
em->generation = -1;
while (1) {
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- if (!ret)
- list_move(&em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
@@ -914,11 +951,14 @@ static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
btrfs_drop_extent_cache(inode, start,
start + ram_size - 1, 0);
}
+ if (ret)
+ goto out_reserve;
cur_alloc_size = ins.offset;
ret = btrfs_add_ordered_extent(inode, start, ins.objectid,
ram_size, cur_alloc_size, 0);
- BUG_ON(ret); /* -ENOMEM */
+ if (ret)
+ goto out_reserve;
if (root->root_key.objectid ==
BTRFS_DATA_RELOC_TREE_OBJECTID) {
@@ -926,7 +966,7 @@ static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
cur_alloc_size);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
- goto out_unlock;
+ goto out_reserve;
}
}
@@ -955,6 +995,8 @@ static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
out:
return ret;
+out_reserve:
+ btrfs_free_reserved_extent(root, ins.objectid, ins.offset);
out_unlock:
extent_clear_unlock_delalloc(inode,
&BTRFS_I(inode)->io_tree,
@@ -1158,6 +1200,7 @@ static noinline int run_delalloc_nocow(struct inode *inode,
u64 disk_bytenr;
u64 num_bytes;
u64 disk_num_bytes;
+ u64 ram_bytes;
int extent_type;
int ret, err;
int type;
@@ -1254,6 +1297,7 @@ next_slot:
struct btrfs_file_extent_item);
extent_type = btrfs_file_extent_type(leaf, fi);
+ ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
if (extent_type == BTRFS_FILE_EXTENT_REG ||
extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
@@ -1337,16 +1381,16 @@ out_check:
em->block_len = num_bytes;
em->block_start = disk_bytenr;
em->orig_block_len = disk_num_bytes;
+ em->ram_bytes = ram_bytes;
em->bdev = root->fs_info->fs_devices->latest_bdev;
+ em->mod_start = em->start;
+ em->mod_len = em->len;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
set_bit(EXTENT_FLAG_FILLING, &em->flags);
em->generation = -1;
while (1) {
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- if (!ret)
- list_move(&em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
@@ -1487,7 +1531,7 @@ static void btrfs_merge_extent_hook(struct inode *inode,
* have pending delalloc work to be done.
*/
static void btrfs_set_bit_hook(struct inode *inode,
- struct extent_state *state, int *bits)
+ struct extent_state *state, unsigned long *bits)
{
/*
@@ -1508,14 +1552,22 @@ static void btrfs_set_bit_hook(struct inode *inode,
spin_unlock(&BTRFS_I(inode)->lock);
}
- spin_lock(&root->fs_info->delalloc_lock);
+ __percpu_counter_add(&root->fs_info->delalloc_bytes, len,
+ root->fs_info->delalloc_batch);
+ spin_lock(&BTRFS_I(inode)->lock);
BTRFS_I(inode)->delalloc_bytes += len;
- root->fs_info->delalloc_bytes += len;
- if (do_list && list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
- list_add_tail(&BTRFS_I(inode)->delalloc_inodes,
- &root->fs_info->delalloc_inodes);
+ if (do_list && !test_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &BTRFS_I(inode)->runtime_flags)) {
+ spin_lock(&root->fs_info->delalloc_lock);
+ if (list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
+ list_add_tail(&BTRFS_I(inode)->delalloc_inodes,
+ &root->fs_info->delalloc_inodes);
+ set_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &BTRFS_I(inode)->runtime_flags);
+ }
+ spin_unlock(&root->fs_info->delalloc_lock);
}
- spin_unlock(&root->fs_info->delalloc_lock);
+ spin_unlock(&BTRFS_I(inode)->lock);
}
}
@@ -1523,7 +1575,8 @@ static void btrfs_set_bit_hook(struct inode *inode,
* extent_io.c clear_bit_hook, see set_bit_hook for why
*/
static void btrfs_clear_bit_hook(struct inode *inode,
- struct extent_state *state, int *bits)
+ struct extent_state *state,
+ unsigned long *bits)
{
/*
* set_bit and clear bit hooks normally require _irqsave/restore
@@ -1550,15 +1603,22 @@ static void btrfs_clear_bit_hook(struct inode *inode,
&& do_list)
btrfs_free_reserved_data_space(inode, len);
- spin_lock(&root->fs_info->delalloc_lock);
- root->fs_info->delalloc_bytes -= len;
+ __percpu_counter_add(&root->fs_info->delalloc_bytes, -len,
+ root->fs_info->delalloc_batch);
+ spin_lock(&BTRFS_I(inode)->lock);
BTRFS_I(inode)->delalloc_bytes -= len;
-
if (do_list && BTRFS_I(inode)->delalloc_bytes == 0 &&
- !list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
- list_del_init(&BTRFS_I(inode)->delalloc_inodes);
+ test_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &BTRFS_I(inode)->runtime_flags)) {
+ spin_lock(&root->fs_info->delalloc_lock);
+ if (!list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
+ list_del_init(&BTRFS_I(inode)->delalloc_inodes);
+ clear_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &BTRFS_I(inode)->runtime_flags);
+ }
+ spin_unlock(&root->fs_info->delalloc_lock);
}
- spin_unlock(&root->fs_info->delalloc_lock);
+ spin_unlock(&BTRFS_I(inode)->lock);
}
}
@@ -1566,7 +1626,7 @@ static void btrfs_clear_bit_hook(struct inode *inode,
* extent_io.c merge_bio_hook, this must check the chunk tree to make sure
* we don't create bios that span stripes or chunks
*/
-int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
+int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
size_t size, struct bio *bio,
unsigned long bio_flags)
{
@@ -1581,7 +1641,7 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
length = bio->bi_size;
map_length = length;
- ret = btrfs_map_block(root->fs_info, READ, logical,
+ ret = btrfs_map_block(root->fs_info, rw, logical,
&map_length, NULL, 0);
/* Will always return 0 with map_multi == NULL */
BUG_ON(ret < 0);
@@ -1704,8 +1764,10 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
struct btrfs_ordered_sum *sum;
list_for_each_entry(sum, list, list) {
+ trans->adding_csums = 1;
btrfs_csum_file_blocks(trans,
BTRFS_I(inode)->root->fs_info->csum_root, sum);
+ trans->adding_csums = 0;
}
return 0;
}
@@ -1892,6 +1954,643 @@ out:
return ret;
}
+/* snapshot-aware defrag */
+struct sa_defrag_extent_backref {
+ struct rb_node node;
+ struct old_sa_defrag_extent *old;
+ u64 root_id;
+ u64 inum;
+ u64 file_pos;
+ u64 extent_offset;
+ u64 num_bytes;
+ u64 generation;
+};
+
+struct old_sa_defrag_extent {
+ struct list_head list;
+ struct new_sa_defrag_extent *new;
+
+ u64 extent_offset;
+ u64 bytenr;
+ u64 offset;
+ u64 len;
+ int count;
+};
+
+struct new_sa_defrag_extent {
+ struct rb_root root;
+ struct list_head head;
+ struct btrfs_path *path;
+ struct inode *inode;
+ u64 file_pos;
+ u64 len;
+ u64 bytenr;
+ u64 disk_len;
+ u8 compress_type;
+};
+
+static int backref_comp(struct sa_defrag_extent_backref *b1,
+ struct sa_defrag_extent_backref *b2)
+{
+ if (b1->root_id < b2->root_id)
+ return -1;
+ else if (b1->root_id > b2->root_id)
+ return 1;
+
+ if (b1->inum < b2->inum)
+ return -1;
+ else if (b1->inum > b2->inum)
+ return 1;
+
+ if (b1->file_pos < b2->file_pos)
+ return -1;
+ else if (b1->file_pos > b2->file_pos)
+ return 1;
+
+ /*
+ * [------------------------------] ===> (a range of space)
+ * |<--->| |<---->| =============> (fs/file tree A)
+ * |<---------------------------->| ===> (fs/file tree B)
+ *
+ * A range of space can refer to two file extents in one tree while
+ * refer to only one file extent in another tree.
+ *
+ * So we may process a disk offset more than one time(two extents in A)
+ * and locate at the same extent(one extent in B), then insert two same
+ * backrefs(both refer to the extent in B).
+ */
+ return 0;
+}
+
+static void backref_insert(struct rb_root *root,
+ struct sa_defrag_extent_backref *backref)
+{
+ struct rb_node **p = &root->rb_node;
+ struct rb_node *parent = NULL;
+ struct sa_defrag_extent_backref *entry;
+ int ret;
+
+ while (*p) {
+ parent = *p;
+ entry = rb_entry(parent, struct sa_defrag_extent_backref, node);
+
+ ret = backref_comp(backref, entry);
+ if (ret < 0)
+ p = &(*p)->rb_left;
+ else
+ p = &(*p)->rb_right;
+ }
+
+ rb_link_node(&backref->node, parent, p);
+ rb_insert_color(&backref->node, root);
+}
+
+/*
+ * Note the backref might has changed, and in this case we just return 0.
+ */
+static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
+ void *ctx)
+{
+ struct btrfs_file_extent_item *extent;
+ struct btrfs_fs_info *fs_info;
+ struct old_sa_defrag_extent *old = ctx;
+ struct new_sa_defrag_extent *new = old->new;
+ struct btrfs_path *path = new->path;
+ struct btrfs_key key;
+ struct btrfs_root *root;
+ struct sa_defrag_extent_backref *backref;
+ struct extent_buffer *leaf;
+ struct inode *inode = new->inode;
+ int slot;
+ int ret;
+ u64 extent_offset;
+ u64 num_bytes;
+
+ if (BTRFS_I(inode)->root->root_key.objectid == root_id &&
+ inum == btrfs_ino(inode))
+ return 0;
+
+ key.objectid = root_id;
+ key.type = BTRFS_ROOT_ITEM_KEY;
+ key.offset = (u64)-1;
+
+ fs_info = BTRFS_I(inode)->root->fs_info;
+ root = btrfs_read_fs_root_no_name(fs_info, &key);
+ if (IS_ERR(root)) {
+ if (PTR_ERR(root) == -ENOENT)
+ return 0;
+ WARN_ON(1);
+ pr_debug("inum=%llu, offset=%llu, root_id=%llu\n",
+ inum, offset, root_id);
+ return PTR_ERR(root);
+ }
+
+ key.objectid = inum;
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ if (offset > (u64)-1 << 32)
+ key.offset = 0;
+ else
+ key.offset = offset;
+
+ ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+ if (ret < 0) {
+ WARN_ON(1);
+ return ret;
+ }
+
+ while (1) {
+ cond_resched();
+
+ leaf = path->nodes[0];
+ slot = path->slots[0];
+
+ if (slot >= btrfs_header_nritems(leaf)) {
+ ret = btrfs_next_leaf(root, path);
+ if (ret < 0) {
+ goto out;
+ } else if (ret > 0) {
+ ret = 0;
+ goto out;
+ }
+ continue;
+ }
+
+ path->slots[0]++;
+
+ btrfs_item_key_to_cpu(leaf, &key, slot);
+
+ if (key.objectid > inum)
+ goto out;
+
+ if (key.objectid < inum || key.type != BTRFS_EXTENT_DATA_KEY)
+ continue;
+
+ extent = btrfs_item_ptr(leaf, slot,
+ struct btrfs_file_extent_item);
+
+ if (btrfs_file_extent_disk_bytenr(leaf, extent) != old->bytenr)
+ continue;
+
+ extent_offset = btrfs_file_extent_offset(leaf, extent);
+ if (key.offset - extent_offset != offset)
+ continue;
+
+ num_bytes = btrfs_file_extent_num_bytes(leaf, extent);
+ if (extent_offset >= old->extent_offset + old->offset +
+ old->len || extent_offset + num_bytes <=
+ old->extent_offset + old->offset)
+ continue;
+
+ break;
+ }
+
+ backref = kmalloc(sizeof(*backref), GFP_NOFS);
+ if (!backref) {
+ ret = -ENOENT;
+ goto out;
+ }
+
+ backref->root_id = root_id;
+ backref->inum = inum;
+ backref->file_pos = offset + extent_offset;
+ backref->num_bytes = num_bytes;
+ backref->extent_offset = extent_offset;
+ backref->generation = btrfs_file_extent_generation(leaf, extent);
+ backref->old = old;
+ backref_insert(&new->root, backref);
+ old->count++;
+out:
+ btrfs_release_path(path);
+ WARN_ON(ret);
+ return ret;
+}
+
+static noinline bool record_extent_backrefs(struct btrfs_path *path,
+ struct new_sa_defrag_extent *new)
+{
+ struct btrfs_fs_info *fs_info = BTRFS_I(new->inode)->root->fs_info;
+ struct old_sa_defrag_extent *old, *tmp;
+ int ret;
+
+ new->path = path;
+
+ list_for_each_entry_safe(old, tmp, &new->head, list) {
+ ret = iterate_inodes_from_logical(old->bytenr, fs_info,
+ path, record_one_backref,
+ old);
+ BUG_ON(ret < 0 && ret != -ENOENT);
+
+ /* no backref to be processed for this extent */
+ if (!old->count) {
+ list_del(&old->list);
+ kfree(old);
+ }
+ }
+
+ if (list_empty(&new->head))
+ return false;
+
+ return true;
+}
+
+static int relink_is_mergable(struct extent_buffer *leaf,
+ struct btrfs_file_extent_item *fi,
+ u64 disk_bytenr)
+{
+ if (btrfs_file_extent_disk_bytenr(leaf, fi) != disk_bytenr)
+ return 0;
+
+ if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG)
+ return 0;
+
+ if (btrfs_file_extent_compression(leaf, fi) ||
+ btrfs_file_extent_encryption(leaf, fi) ||
+ btrfs_file_extent_other_encoding(leaf, fi))
+ return 0;
+
+ return 1;
+}
+
+/*
+ * Note the backref might has changed, and in this case we just return 0.
+ */
+static noinline int relink_extent_backref(struct btrfs_path *path,
+ struct sa_defrag_extent_backref *prev,
+ struct sa_defrag_extent_backref *backref)
+{
+ struct btrfs_file_extent_item *extent;
+ struct btrfs_file_extent_item *item;
+ struct btrfs_ordered_extent *ordered;
+ struct btrfs_trans_handle *trans;
+ struct btrfs_fs_info *fs_info;
+ struct btrfs_root *root;
+ struct btrfs_key key;
+ struct extent_buffer *leaf;
+ struct old_sa_defrag_extent *old = backref->old;
+ struct new_sa_defrag_extent *new = old->new;
+ struct inode *src_inode = new->inode;
+ struct inode *inode;
+ struct extent_state *cached = NULL;
+ int ret = 0;
+ u64 start;
+ u64 len;
+ u64 lock_start;
+ u64 lock_end;
+ bool merge = false;
+ int index;
+
+ if (prev && prev->root_id == backref->root_id &&
+ prev->inum == backref->inum &&
+ prev->file_pos + prev->num_bytes == backref->file_pos)
+ merge = true;
+
+ /* step 1: get root */
+ key.objectid = backref->root_id;
+ key.type = BTRFS_ROOT_ITEM_KEY;
+ key.offset = (u64)-1;
+
+ fs_info = BTRFS_I(src_inode)->root->fs_info;
+ index = srcu_read_lock(&fs_info->subvol_srcu);
+
+ root = btrfs_read_fs_root_no_name(fs_info, &key);
+ if (IS_ERR(root)) {
+ srcu_read_unlock(&fs_info->subvol_srcu, index);
+ if (PTR_ERR(root) == -ENOENT)
+ return 0;
+ return PTR_ERR(root);
+ }
+ if (btrfs_root_refs(&root->root_item) == 0) {
+ srcu_read_unlock(&fs_info->subvol_srcu, index);
+ /* parse ENOENT to 0 */
+ return 0;
+ }
+
+ /* step 2: get inode */
+ key.objectid = backref->inum;
+ key.type = BTRFS_INODE_ITEM_KEY;
+ key.offset = 0;
+
+ inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+ if (IS_ERR(inode)) {
+ srcu_read_unlock(&fs_info->subvol_srcu, index);
+ return 0;
+ }
+
+ srcu_read_unlock(&fs_info->subvol_srcu, index);
+
+ /* step 3: relink backref */
+ lock_start = backref->file_pos;
+ lock_end = backref->file_pos + backref->num_bytes - 1;
+ lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
+ 0, &cached);
+
+ ordered = btrfs_lookup_first_ordered_extent(inode, lock_end);
+ if (ordered) {
+ btrfs_put_ordered_extent(ordered);
+ goto out_unlock;
+ }
+
+ trans = btrfs_join_transaction(root);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out_unlock;
+ }
+
+ key.objectid = backref->inum;
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = backref->file_pos;
+
+ ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+ if (ret < 0) {
+ goto out_free_path;
+ } else if (ret > 0) {
+ ret = 0;
+ goto out_free_path;
+ }
+
+ extent = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_file_extent_item);
+
+ if (btrfs_file_extent_generation(path->nodes[0], extent) !=
+ backref->generation)
+ goto out_free_path;
+
+ btrfs_release_path(path);
+
+ start = backref->file_pos;
+ if (backref->extent_offset < old->extent_offset + old->offset)
+ start += old->extent_offset + old->offset -
+ backref->extent_offset;
+
+ len = min(backref->extent_offset + backref->num_bytes,
+ old->extent_offset + old->offset + old->len);
+ len -= max(backref->extent_offset, old->extent_offset + old->offset);
+
+ ret = btrfs_drop_extents(trans, root, inode, start,
+ start + len, 1);
+ if (ret)
+ goto out_free_path;
+again:
+ key.objectid = btrfs_ino(inode);
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = start;
+
+ path->leave_spinning = 1;
+ if (merge) {
+ struct btrfs_file_extent_item *fi;
+ u64 extent_len;
+ struct btrfs_key found_key;
+
+ ret = btrfs_search_slot(trans, root, &key, path, 1, 1);
+ if (ret < 0)
+ goto out_free_path;
+
+ path->slots[0]--;
+ leaf = path->nodes[0];
+ btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
+
+ fi = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_file_extent_item);
+ extent_len = btrfs_file_extent_num_bytes(leaf, fi);
+
+ if (relink_is_mergable(leaf, fi, new->bytenr) &&
+ extent_len + found_key.offset == start) {
+ btrfs_set_file_extent_num_bytes(leaf, fi,
+ extent_len + len);
+ btrfs_mark_buffer_dirty(leaf);
+ inode_add_bytes(inode, len);
+
+ ret = 1;
+ goto out_free_path;
+ } else {
+ merge = false;
+ btrfs_release_path(path);
+ goto again;
+ }
+ }
+
+ ret = btrfs_insert_empty_item(trans, root, path, &key,
+ sizeof(*extent));
+ if (ret) {
+ btrfs_abort_transaction(trans, root, ret);
+ goto out_free_path;
+ }
+
+ leaf = path->nodes[0];
+ item = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_file_extent_item);
+ btrfs_set_file_extent_disk_bytenr(leaf, item, new->bytenr);
+ btrfs_set_file_extent_disk_num_bytes(leaf, item, new->disk_len);
+ btrfs_set_file_extent_offset(leaf, item, start - new->file_pos);
+ btrfs_set_file_extent_num_bytes(leaf, item, len);
+ btrfs_set_file_extent_ram_bytes(leaf, item, new->len);
+ btrfs_set_file_extent_generation(leaf, item, trans->transid);
+ btrfs_set_file_extent_type(leaf, item, BTRFS_FILE_EXTENT_REG);
+ btrfs_set_file_extent_compression(leaf, item, new->compress_type);
+ btrfs_set_file_extent_encryption(leaf, item, 0);
+ btrfs_set_file_extent_other_encoding(leaf, item, 0);
+
+ btrfs_mark_buffer_dirty(leaf);
+ inode_add_bytes(inode, len);
+ btrfs_release_path(path);
+
+ ret = btrfs_inc_extent_ref(trans, root, new->bytenr,
+ new->disk_len, 0,
+ backref->root_id, backref->inum,
+ new->file_pos, 0); /* start - extent_offset */
+ if (ret) {
+ btrfs_abort_transaction(trans, root, ret);
+ goto out_free_path;
+ }
+
+ ret = 1;
+out_free_path:
+ btrfs_release_path(path);
+ path->leave_spinning = 0;
+ btrfs_end_transaction(trans, root);
+out_unlock:
+ unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
+ &cached, GFP_NOFS);
+ iput(inode);
+ return ret;
+}
+
+static void relink_file_extents(struct new_sa_defrag_extent *new)
+{
+ struct btrfs_path *path;
+ struct old_sa_defrag_extent *old, *tmp;
+ struct sa_defrag_extent_backref *backref;
+ struct sa_defrag_extent_backref *prev = NULL;
+ struct inode *inode;
+ struct btrfs_root *root;
+ struct rb_node *node;
+ int ret;
+
+ inode = new->inode;
+ root = BTRFS_I(inode)->root;
+
+ path = btrfs_alloc_path();
+ if (!path)
+ return;
+
+ if (!record_extent_backrefs(path, new)) {
+ btrfs_free_path(path);
+ goto out;
+ }
+ btrfs_release_path(path);
+
+ while (1) {
+ node = rb_first(&new->root);
+ if (!node)
+ break;
+ rb_erase(node, &new->root);
+
+ backref = rb_entry(node, struct sa_defrag_extent_backref, node);
+
+ ret = relink_extent_backref(path, prev, backref);
+ WARN_ON(ret < 0);
+
+ kfree(prev);
+
+ if (ret == 1)
+ prev = backref;
+ else
+ prev = NULL;
+ cond_resched();
+ }
+ kfree(prev);
+
+ btrfs_free_path(path);
+
+ list_for_each_entry_safe(old, tmp, &new->head, list) {
+ list_del(&old->list);
+ kfree(old);
+ }
+out:
+ atomic_dec(&root->fs_info->defrag_running);
+ wake_up(&root->fs_info->transaction_wait);
+
+ kfree(new);
+}
+
+static struct new_sa_defrag_extent *
+record_old_file_extents(struct inode *inode,
+ struct btrfs_ordered_extent *ordered)
+{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_path *path;
+ struct btrfs_key key;
+ struct old_sa_defrag_extent *old, *tmp;
+ struct new_sa_defrag_extent *new;
+ int ret;
+
+ new = kmalloc(sizeof(*new), GFP_NOFS);
+ if (!new)
+ return NULL;
+
+ new->inode = inode;
+ new->file_pos = ordered->file_offset;
+ new->len = ordered->len;
+ new->bytenr = ordered->start;
+ new->disk_len = ordered->disk_len;
+ new->compress_type = ordered->compress_type;
+ new->root = RB_ROOT;
+ INIT_LIST_HEAD(&new->head);
+
+ path = btrfs_alloc_path();
+ if (!path)
+ goto out_kfree;
+
+ key.objectid = btrfs_ino(inode);
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = new->file_pos;
+
+ ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+ if (ret < 0)
+ goto out_free_path;
+ if (ret > 0 && path->slots[0] > 0)
+ path->slots[0]--;
+
+ /* find out all the old extents for the file range */
+ while (1) {
+ struct btrfs_file_extent_item *extent;
+ struct extent_buffer *l;
+ int slot;
+ u64 num_bytes;
+ u64 offset;
+ u64 end;
+ u64 disk_bytenr;
+ u64 extent_offset;
+
+ l = path->nodes[0];
+ slot = path->slots[0];
+
+ if (slot >= btrfs_header_nritems(l)) {
+ ret = btrfs_next_leaf(root, path);
+ if (ret < 0)
+ goto out_free_list;
+ else if (ret > 0)
+ break;
+ continue;
+ }
+
+ btrfs_item_key_to_cpu(l, &key, slot);
+
+ if (key.objectid != btrfs_ino(inode))
+ break;
+ if (key.type != BTRFS_EXTENT_DATA_KEY)
+ break;
+ if (key.offset >= new->file_pos + new->len)
+ break;
+
+ extent = btrfs_item_ptr(l, slot, struct btrfs_file_extent_item);
+
+ num_bytes = btrfs_file_extent_num_bytes(l, extent);
+ if (key.offset + num_bytes < new->file_pos)
+ goto next;
+
+ disk_bytenr = btrfs_file_extent_disk_bytenr(l, extent);
+ if (!disk_bytenr)
+ goto next;
+
+ extent_offset = btrfs_file_extent_offset(l, extent);
+
+ old = kmalloc(sizeof(*old), GFP_NOFS);
+ if (!old)
+ goto out_free_list;
+
+ offset = max(new->file_pos, key.offset);
+ end = min(new->file_pos + new->len, key.offset + num_bytes);
+
+ old->bytenr = disk_bytenr;
+ old->extent_offset = extent_offset;
+ old->offset = offset - key.offset;
+ old->len = end - offset;
+ old->new = new;
+ old->count = 0;
+ list_add_tail(&old->list, &new->head);
+next:
+ path->slots[0]++;
+ cond_resched();
+ }
+
+ btrfs_free_path(path);
+ atomic_inc(&root->fs_info->defrag_running);
+
+ return new;
+
+out_free_list:
+ list_for_each_entry_safe(old, tmp, &new->head, list) {
+ list_del(&old->list);
+ kfree(old);
+ }
+out_free_path:
+ btrfs_free_path(path);
+out_kfree:
+ kfree(new);
+ return NULL;
+}
+
/*
* helper function for btrfs_finish_ordered_io, this
* just reads in some of the csum leaves to prime them into ram
@@ -1909,6 +2608,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
struct btrfs_trans_handle *trans = NULL;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
struct extent_state *cached_state = NULL;
+ struct new_sa_defrag_extent *new = NULL;
int compress_type = 0;
int ret;
bool nolock;
@@ -1943,6 +2643,20 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
ordered_extent->file_offset + ordered_extent->len - 1,
0, &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);
+ if (ret) {
+ u64 last_snapshot = btrfs_root_last_snapshot(&root->root_item);
+ if (last_snapshot >= BTRFS_I(inode)->generation)
+ /* the inode is shared */
+ new = record_old_file_extents(inode, ordered_extent);
+
+ clear_extent_bit(io_tree, ordered_extent->file_offset,
+ ordered_extent->file_offset + ordered_extent->len - 1,
+ EXTENT_DEFRAG, 0, 0, &cached_state, GFP_NOFS);
+ }
+
if (nolock)
trans = btrfs_join_transaction_nolock(root);
else
@@ -2001,17 +2715,33 @@ out:
if (trans)
btrfs_end_transaction(trans, root);
- if (ret)
+ if (ret) {
clear_extent_uptodate(io_tree, ordered_extent->file_offset,
ordered_extent->file_offset +
ordered_extent->len - 1, NULL, GFP_NOFS);
+ /*
+ * If the ordered extent had an IOERR or something else went
+ * wrong we need to return the space for this ordered extent
+ * back to the allocator.
+ */
+ if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
+ !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags))
+ btrfs_free_reserved_extent(root, ordered_extent->start,
+ ordered_extent->disk_len);
+ }
+
+
/*
* This needs to be done to make sure anybody waiting knows we are done
* updating everything for this ordered extent.
*/
btrfs_remove_ordered_extent(inode, ordered_extent);
+ /* for snapshot-aware defrag */
+ if (new)
+ relink_file_extents(new);
+
/* once for us */
btrfs_put_ordered_extent(ordered_extent);
/* once for the tree */
@@ -2062,7 +2792,7 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state, int mirror)
{
- size_t offset = start - ((u64)page->index << PAGE_CACHE_SHIFT);
+ size_t offset = start - page_offset(page);
struct inode *inode = page->mapping->host;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
char *kaddr;
@@ -2070,6 +2800,8 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
int ret;
struct btrfs_root *root = BTRFS_I(inode)->root;
u32 csum = ~(u32)0;
+ static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL,
+ DEFAULT_RATELIMIT_BURST);
if (PageChecked(page)) {
ClearPageChecked(page);
@@ -2096,7 +2828,7 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
if (ret)
goto zeroit;
- csum = btrfs_csum_data(root, kaddr + offset, csum, end - start + 1);
+ csum = btrfs_csum_data(kaddr + offset, csum, end - start + 1);
btrfs_csum_final(csum, (char *)&csum);
if (csum != private)
goto zeroit;
@@ -2106,11 +2838,11 @@ good:
return 0;
zeroit:
- printk_ratelimited(KERN_INFO "btrfs csum failed ino %llu off %llu csum %u "
- "private %llu\n",
- (unsigned long long)btrfs_ino(page->mapping->host),
- (unsigned long long)start, csum,
- (unsigned long long)private);
+ if (__ratelimit(&_rs))
+ btrfs_info(root->fs_info, "csum failed ino %llu off %llu csum %u private %llu",
+ (unsigned long long)btrfs_ino(page->mapping->host),
+ (unsigned long long)start, csum,
+ (unsigned long long)private);
memset(kaddr + offset, 1, end - start + 1);
flush_dcache_page(page);
kunmap_atomic(kaddr);
@@ -2167,11 +2899,6 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root)
}
}
-enum btrfs_orphan_cleanup_state {
- ORPHAN_CLEANUP_STARTED = 1,
- ORPHAN_CLEANUP_DONE = 2,
-};
-
/*
* This is called in transaction commit time. If there are no orphan
* files in the subvolume, it removes orphan item and frees block_rsv
@@ -2301,7 +3028,8 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
* We have done the truncate/delete so we can go ahead and remove the orphan
* item for this particular inode.
*/
-int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct inode *inode)
+static int btrfs_orphan_del(struct btrfs_trans_handle *trans,
+ struct inode *inode)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
int delete_item = 0;
@@ -2396,8 +3124,8 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
*/
if (found_key.offset == last_objectid) {
- printk(KERN_ERR "btrfs: Error removing orphan entry, "
- "stopping orphan cleanup\n");
+ btrfs_err(root->fs_info,
+ "Error removing orphan entry, stopping orphan cleanup");
ret = -EINVAL;
goto out;
}
@@ -2454,8 +3182,8 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
ret = PTR_ERR(trans);
goto out;
}
- printk(KERN_ERR "auto deleting %Lu\n",
- found_key.objectid);
+ btrfs_debug(root->fs_info, "auto deleting %Lu",
+ found_key.objectid);
ret = btrfs_del_orphan_item(trans, root,
found_key.objectid);
BUG_ON(ret); /* -ENOMEM or corruption (JDM: Recheck) */
@@ -2469,6 +3197,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
*/
set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
&BTRFS_I(inode)->runtime_flags);
+ atomic_inc(&root->orphan_inodes);
/* if we have links, this was a truncate, lets do that */
if (inode->i_nlink) {
@@ -2491,6 +3220,8 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
goto out;
ret = btrfs_truncate(inode);
+ if (ret)
+ btrfs_orphan_del(NULL, inode);
} else {
nr_unlink++;
}
@@ -2516,13 +3247,14 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
}
if (nr_unlink)
- printk(KERN_INFO "btrfs: unlinked %d orphans\n", nr_unlink);
+ btrfs_debug(root->fs_info, "unlinked %d orphans", nr_unlink);
if (nr_truncate)
- printk(KERN_INFO "btrfs: truncated %d orphans\n", nr_truncate);
+ btrfs_debug(root->fs_info, "truncated %d orphans", nr_truncate);
out:
if (ret)
- printk(KERN_CRIT "btrfs: could not do orphan cleanup %d\n", ret);
+ btrfs_crit(root->fs_info,
+ "could not do orphan cleanup %d", ret);
btrfs_free_path(path);
return ret;
}
@@ -2709,34 +3441,41 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
struct btrfs_inode_item *item,
struct inode *inode)
{
- btrfs_set_inode_uid(leaf, item, i_uid_read(inode));
- btrfs_set_inode_gid(leaf, item, i_gid_read(inode));
- btrfs_set_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size);
- btrfs_set_inode_mode(leaf, item, inode->i_mode);
- btrfs_set_inode_nlink(leaf, item, inode->i_nlink);
+ struct btrfs_map_token token;
+
+ btrfs_init_map_token(&token);
+
+ btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token);
+ btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token);
+ btrfs_set_token_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size,
+ &token);
+ btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token);
+ btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_atime(item),
- inode->i_atime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_atime(item),
- inode->i_atime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_atime(item),
+ inode->i_atime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_atime(item),
+ inode->i_atime.tv_nsec, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_mtime(item),
- inode->i_mtime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_mtime(item),
- inode->i_mtime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_mtime(item),
+ inode->i_mtime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_mtime(item),
+ inode->i_mtime.tv_nsec, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_ctime(item),
- inode->i_ctime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_ctime(item),
- inode->i_ctime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_ctime(item),
+ inode->i_ctime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_ctime(item),
+ inode->i_ctime.tv_nsec, &token);
- btrfs_set_inode_nbytes(leaf, item, inode_get_bytes(inode));
- btrfs_set_inode_generation(leaf, item, BTRFS_I(inode)->generation);
- btrfs_set_inode_sequence(leaf, item, inode->i_version);
- btrfs_set_inode_transid(leaf, item, trans->transid);
- btrfs_set_inode_rdev(leaf, item, inode->i_rdev);
- btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags);
- btrfs_set_inode_block_group(leaf, item, 0);
+ btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode),
+ &token);
+ btrfs_set_token_inode_generation(leaf, item, BTRFS_I(inode)->generation,
+ &token);
+ btrfs_set_token_inode_sequence(leaf, item, inode->i_version, &token);
+ btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
+ btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
+ btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
+ btrfs_set_token_inode_block_group(leaf, item, 0, &token);
}
/*
@@ -2863,9 +3602,10 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
ret = btrfs_del_inode_ref(trans, root, name, name_len, ino,
dir_ino, &index);
if (ret) {
- printk(KERN_INFO "btrfs failed to delete reference to %.*s, "
- "inode %llu parent %llu\n", name_len, name,
- (unsigned long long)ino, (unsigned long long)dir_ino);
+ btrfs_info(root->fs_info,
+ "failed to delete reference to %.*s, inode %llu parent %llu",
+ name_len, name,
+ (unsigned long long)ino, (unsigned long long)dir_ino);
btrfs_abort_transaction(trans, root, ret);
goto err;
}
@@ -2887,6 +3627,8 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
dir, index);
if (ret == -ENOENT)
ret = 0;
+ else if (ret)
+ btrfs_abort_transaction(trans, root, ret);
err:
btrfs_free_path(path);
if (ret)
@@ -2932,7 +3674,7 @@ static int check_path_shared(struct btrfs_root *root,
eb = path->nodes[level];
if (!btrfs_block_can_be_shared(root, eb))
continue;
- ret = btrfs_lookup_extent_info(NULL, root, eb->start, eb->len,
+ ret = btrfs_lookup_extent_info(NULL, root, eb->start, level, 1,
&refs, NULL);
if (refs > 1)
return 1;
@@ -2967,11 +3709,9 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
* 1 for the dir item
* 1 for the dir index
* 1 for the inode ref
- * 1 for the inode ref in the tree log
- * 2 for the dir entries in the log
* 1 for the inode
*/
- trans = btrfs_start_transaction(root, 8);
+ trans = btrfs_start_transaction(root, 5);
if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
return trans;
@@ -3304,7 +4044,6 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
u64 extent_num_bytes = 0;
u64 extent_offset = 0;
u64 item_end = 0;
- u64 mask = root->sectorsize - 1;
u32 found_type = (u8)-1;
int found_extent;
int del_item;
@@ -3328,7 +4067,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
* extent just the way it is.
*/
if (root->ref_cows || root == root->fs_info->tree_root)
- btrfs_drop_extent_cache(inode, (new_size + mask) & (~mask), (u64)-1, 0);
+ btrfs_drop_extent_cache(inode, ALIGN(new_size,
+ root->sectorsize), (u64)-1, 0);
/*
* This function is also used to drop the items in the log tree before
@@ -3407,10 +4147,9 @@ search_again:
if (!del_item) {
u64 orig_num_bytes =
btrfs_file_extent_num_bytes(leaf, fi);
- extent_num_bytes = new_size -
- found_key.offset + root->sectorsize - 1;
- extent_num_bytes = extent_num_bytes &
- ~((u64)root->sectorsize - 1);
+ extent_num_bytes = ALIGN(new_size -
+ found_key.offset,
+ root->sectorsize);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_num_bytes);
num_dec = (orig_num_bytes -
@@ -3450,8 +4189,7 @@ search_again:
}
size =
btrfs_file_extent_calc_inline_size(size);
- btrfs_truncate_item(trans, root, path,
- size, 1);
+ btrfs_truncate_item(root, path, size, 1);
} else if (root->ref_cows) {
inode_sub_bytes(inode, item_end + 1 -
found_key.offset);
@@ -3646,9 +4384,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
struct extent_map *em = NULL;
struct extent_state *cached_state = NULL;
struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
- u64 mask = root->sectorsize - 1;
- u64 hole_start = (oldsize + mask) & ~mask;
- u64 block_end = (size + mask) & ~mask;
+ u64 hole_start = ALIGN(oldsize, root->sectorsize);
+ u64 block_end = ALIGN(size, root->sectorsize);
u64 last_byte;
u64 cur_offset;
u64 hole_size;
@@ -3681,7 +4418,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
break;
}
last_byte = min(extent_map_end(em), block_end);
- last_byte = (last_byte + mask) & ~mask;
+ last_byte = ALIGN(last_byte , root->sectorsize);
if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
struct extent_map *hole_em;
hole_size = last_byte - cur_offset;
@@ -3726,16 +4463,14 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
hole_em->block_start = EXTENT_MAP_HOLE;
hole_em->block_len = 0;
hole_em->orig_block_len = 0;
+ hole_em->ram_bytes = hole_size;
hole_em->bdev = root->fs_info->fs_devices->latest_bdev;
hole_em->compress_type = BTRFS_COMPRESS_NONE;
hole_em->generation = trans->transid;
while (1) {
write_lock(&em_tree->lock);
- err = add_extent_mapping(em_tree, hole_em);
- if (!err)
- list_move(&hole_em->list,
- &em_tree->modified_extents);
+ err = add_extent_mapping(em_tree, hole_em, 1);
write_unlock(&em_tree->lock);
if (err != -EEXIST)
break;
@@ -3832,6 +4567,12 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
/* we don't support swapfiles, so vmtruncate shouldn't fail */
truncate_setsize(inode, newsize);
+
+ /* Disable nonlocked read DIO to avoid the end less truncate */
+ btrfs_inode_block_unlocked_dio(inode);
+ inode_dio_wait(inode);
+ btrfs_inode_resume_unlocked_dio(inode);
+
ret = btrfs_truncate(inode);
if (ret && inode->i_nlink)
btrfs_orphan_del(NULL, inode);
@@ -3904,6 +4645,12 @@ void btrfs_evict_inode(struct inode *inode)
goto no_delete;
}
+ ret = btrfs_commit_inode_delayed_inode(inode);
+ if (ret) {
+ btrfs_orphan_del(NULL, inode);
+ goto no_delete;
+ }
+
rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP);
if (!rsv) {
btrfs_orphan_del(NULL, inode);
@@ -3934,14 +4681,15 @@ void btrfs_evict_inode(struct inode *inode)
ret = btrfs_block_rsv_migrate(global_rsv, rsv, min_size);
if (ret) {
- printk(KERN_WARNING "Could not get space for a "
- "delete, will truncate on mount %d\n", ret);
+ btrfs_warn(root->fs_info,
+ "Could not get space for a delete, will truncate on mount %d",
+ ret);
btrfs_orphan_del(NULL, inode);
btrfs_free_block_rsv(root, rsv);
goto no_delete;
}
- trans = btrfs_start_transaction_lflush(root, 1);
+ trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
btrfs_orphan_del(NULL, inode);
btrfs_free_block_rsv(root, rsv);
@@ -3955,9 +4703,6 @@ void btrfs_evict_inode(struct inode *inode)
break;
trans->block_rsv = &root->fs_info->trans_block_rsv;
- ret = btrfs_update_inode(trans, root, inode);
- BUG_ON(ret);
-
btrfs_end_transaction(trans, root);
trans = NULL;
btrfs_btree_balance_dirty(root);
@@ -4391,7 +5136,7 @@ unsigned char btrfs_filetype_table[] = {
static int btrfs_real_readdir(struct file *filp, void *dirent,
filldir_t filldir)
{
- struct inode *inode = filp->f_dentry->d_inode;
+ struct inode *inode = file_inode(filp);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_item *item;
struct btrfs_dir_item *di;
@@ -4602,7 +5347,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc)
* FIXME, needs more benchmarking...there are no reasons other than performance
* to keep or drop this code.
*/
-int btrfs_dirty_inode(struct inode *inode)
+static int btrfs_dirty_inode(struct inode *inode)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
@@ -4854,7 +5599,8 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
if (btrfs_test_opt(root, NODATASUM))
BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
if (btrfs_test_opt(root, NODATACOW))
- BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW;
+ BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW |
+ BTRFS_INODE_NODATASUM;
}
insert_inode_hash(inode);
@@ -5006,12 +5752,6 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
goto out_unlock;
}
- err = btrfs_update_inode(trans, root, inode);
- if (err) {
- drop_inode = 1;
- goto out_unlock;
- }
-
/*
* If the active LSM wants to access the inode during
* d_instantiate it needs these. Smack checks to see
@@ -5249,7 +5989,7 @@ static int merge_extent_mapping(struct extent_map_tree *em_tree,
em->block_start += start_diff;
em->block_len -= start_diff;
}
- return add_extent_mapping(em_tree, em);
+ return add_extent_mapping(em_tree, em, 0);
}
static noinline int uncompress_inline(struct btrfs_path *path,
@@ -5396,8 +6136,7 @@ again:
} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
size_t size;
size = btrfs_file_extent_inline_len(leaf, item);
- extent_end = (extent_start + size + root->sectorsize - 1) &
- ~((u64)root->sectorsize - 1);
+ extent_end = ALIGN(extent_start + size, root->sectorsize);
}
if (start >= extent_end) {
@@ -5424,6 +6163,7 @@ again:
goto not_found_em;
}
+ em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, item);
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
em->start = extent_start;
@@ -5469,8 +6209,7 @@ again:
copy_size = min_t(u64, PAGE_CACHE_SIZE - pg_offset,
size - extent_offset);
em->start = extent_start + extent_offset;
- em->len = (copy_size + root->sectorsize - 1) &
- ~((u64)root->sectorsize - 1);
+ em->len = ALIGN(copy_size, root->sectorsize);
em->orig_block_len = em->len;
em->orig_start = em->start;
if (compress_type) {
@@ -5533,18 +6272,18 @@ not_found_em:
insert:
btrfs_release_path(path);
if (em->start > start || extent_map_end(em) <= start) {
- printk(KERN_ERR "Btrfs: bad extent! em: [%llu %llu] passed "
- "[%llu %llu]\n", (unsigned long long)em->start,
- (unsigned long long)em->len,
- (unsigned long long)start,
- (unsigned long long)len);
+ btrfs_err(root->fs_info, "bad extent! em: [%llu %llu] passed [%llu %llu]",
+ (unsigned long long)em->start,
+ (unsigned long long)em->len,
+ (unsigned long long)start,
+ (unsigned long long)len);
err = -EIO;
goto out;
}
err = 0;
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
+ ret = add_extent_mapping(em_tree, em, 0);
/* it is possible that someone inserted the extent into the tree
* while we had the lock dropped. It is also possible that
* an overlapping map exists in the tree
@@ -5756,7 +6495,7 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
}
em = create_pinned_em(inode, start, ins.offset, start, ins.objectid,
- ins.offset, ins.offset, 0);
+ ins.offset, ins.offset, ins.offset, 0);
if (IS_ERR(em))
goto out;
@@ -5776,7 +6515,9 @@ out:
* block must be cow'd
*/
static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans,
- struct inode *inode, u64 offset, u64 len)
+ struct inode *inode, u64 offset, u64 *len,
+ u64 *orig_start, u64 *orig_block_len,
+ u64 *ram_bytes)
{
struct btrfs_path *path;
int ret;
@@ -5833,8 +6574,12 @@ static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans,
disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
backref_offset = btrfs_file_extent_offset(leaf, fi);
+ *orig_start = key.offset - backref_offset;
+ *orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi);
+ *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
+
extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
- if (extent_end < offset + len) {
+ if (extent_end < offset + *len) {
/* extent doesn't include our full range, must cow */
goto out;
}
@@ -5858,13 +6603,14 @@ static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans,
*/
disk_bytenr += backref_offset;
disk_bytenr += offset - key.offset;
- num_bytes = min(offset + len, extent_end) - offset;
+ num_bytes = min(offset + *len, extent_end) - offset;
if (csum_exist_in_range(root, disk_bytenr, num_bytes))
goto out;
/*
* all of the above have passed, it is safe to overwrite this extent
* without cow
*/
+ *len = num_bytes;
ret = 1;
out:
btrfs_free_path(path);
@@ -5935,7 +6681,8 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
u64 len, u64 orig_start,
u64 block_start, u64 block_len,
- u64 orig_block_len, int type)
+ u64 orig_block_len, u64 ram_bytes,
+ int type)
{
struct extent_map_tree *em_tree;
struct extent_map *em;
@@ -5949,11 +6696,14 @@ 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;
em->bdev = root->fs_info->fs_devices->latest_bdev;
em->orig_block_len = orig_block_len;
+ em->ram_bytes = ram_bytes;
em->generation = -1;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
if (type == BTRFS_ORDERED_PREALLOC)
@@ -5963,10 +6713,7 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
btrfs_drop_extent_cache(inode, em->start,
em->start + em->len - 1, 0);
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- if (!ret)
- list_move(&em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
} while (ret == -EEXIST);
@@ -5990,16 +6737,12 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
u64 len = bh_result->b_size;
struct btrfs_trans_handle *trans;
int unlock_bits = EXTENT_LOCKED;
- int ret;
+ int ret = 0;
- if (create) {
- ret = btrfs_delalloc_reserve_space(inode, len);
- if (ret)
- return ret;
+ if (create)
unlock_bits |= EXTENT_DELALLOC | EXTENT_DIRTY;
- } else {
+ else
len = min_t(u64, len, root->sectorsize);
- }
lockstart = start;
lockend = start + len - 1;
@@ -6011,14 +6754,6 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
if (lock_extent_direct(inode, lockstart, lockend, &cached_state, create))
return -ENOTBLK;
- if (create) {
- ret = set_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
- lockend, EXTENT_DELALLOC, NULL,
- &cached_state, GFP_NOFS);
- if (ret)
- goto unlock_err;
- }
-
em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
if (IS_ERR(em)) {
ret = PTR_ERR(em);
@@ -6050,7 +6785,6 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
if (!create && (em->block_start == EXTENT_MAP_HOLE ||
test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
free_extent_map(em);
- ret = 0;
goto unlock_err;
}
@@ -6074,7 +6808,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
em->block_start != EXTENT_MAP_HOLE)) {
int type;
int ret;
- u64 block_start;
+ u64 block_start, orig_start, orig_block_len, ram_bytes;
if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
type = BTRFS_ORDERED_PREALLOC;
@@ -6092,16 +6826,15 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
if (IS_ERR(trans))
goto must_cow;
- if (can_nocow_odirect(trans, inode, start, len) == 1) {
- u64 orig_start = em->orig_start;
- u64 orig_block_len = em->orig_block_len;
-
+ if (can_nocow_odirect(trans, inode, start, &len, &orig_start,
+ &orig_block_len, &ram_bytes) == 1) {
if (type == BTRFS_ORDERED_PREALLOC) {
free_extent_map(em);
em = create_pinned_em(inode, start, len,
orig_start,
block_start, len,
- orig_block_len, type);
+ orig_block_len,
+ ram_bytes, type);
if (IS_ERR(em)) {
btrfs_end_transaction(trans, root);
goto unlock_err;
@@ -6148,6 +6881,15 @@ unlock:
*/
if (start + len > i_size_read(inode))
i_size_write(inode, start + len);
+
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->outstanding_extents++;
+ spin_unlock(&BTRFS_I(inode)->lock);
+
+ ret = set_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
+ lockstart + len - 1, EXTENT_DELALLOC, NULL,
+ &cached_state, GFP_NOFS);
+ BUG_ON(ret);
}
/*
@@ -6156,24 +6898,9 @@ unlock:
* aren't using if there is any left over space.
*/
if (lockstart < lockend) {
- if (create && len < lockend - lockstart) {
- clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
- lockstart + len - 1,
- unlock_bits | EXTENT_DEFRAG, 1, 0,
- &cached_state, GFP_NOFS);
- /*
- * Beside unlock, we also need to cleanup reserved space
- * for the left range by attaching EXTENT_DO_ACCOUNTING.
- */
- clear_extent_bit(&BTRFS_I(inode)->io_tree,
- lockstart + len, lockend,
- unlock_bits | EXTENT_DO_ACCOUNTING |
- EXTENT_DEFRAG, 1, 0, NULL, GFP_NOFS);
- } else {
- clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
- lockend, unlock_bits, 1, 0,
- &cached_state, GFP_NOFS);
- }
+ clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
+ lockend, unlock_bits, 1, 0,
+ &cached_state, GFP_NOFS);
} else {
free_extent_state(cached_state);
}
@@ -6183,9 +6910,6 @@ unlock:
return 0;
unlock_err:
- if (create)
- unlock_bits |= EXTENT_DO_ACCOUNTING;
-
clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend,
unlock_bits, 1, 0, &cached_state, GFP_NOFS);
return ret;
@@ -6230,7 +6954,7 @@ static void btrfs_endio_direct_read(struct bio *bio, int err)
goto failed;
local_irq_save(flags);
kaddr = kmap_atomic(page);
- csum = btrfs_csum_data(root, kaddr + bvec->bv_offset,
+ csum = btrfs_csum_data(kaddr + bvec->bv_offset,
csum, bvec->bv_len);
btrfs_csum_final(csum, (char *)&csum);
kunmap_atomic(kaddr);
@@ -6239,11 +6963,10 @@ static void btrfs_endio_direct_read(struct bio *bio, int err)
flush_dcache_page(bvec->bv_page);
if (csum != private) {
failed:
- printk(KERN_ERR "btrfs csum failed ino %llu off"
- " %llu csum %u private %u\n",
- (unsigned long long)btrfs_ino(inode),
- (unsigned long long)start,
- csum, (unsigned)private);
+ btrfs_err(root->fs_info, "csum failed ino %llu off %llu csum %u private %u",
+ (unsigned long long)btrfs_ino(inode),
+ (unsigned long long)start,
+ csum, (unsigned)private);
err = -EIO;
}
}
@@ -6426,19 +7149,24 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
int async_submit = 0;
map_length = orig_bio->bi_size;
- ret = btrfs_map_block(root->fs_info, READ, start_sector << 9,
+ ret = btrfs_map_block(root->fs_info, rw, start_sector << 9,
&map_length, NULL, 0);
if (ret) {
bio_put(orig_bio);
return -EIO;
}
-
if (map_length >= orig_bio->bi_size) {
bio = orig_bio;
goto submit;
}
- async_submit = 1;
+ /* async crcs make it difficult to collect full stripe writes. */
+ if (btrfs_get_alloc_profile(root, 1) &
+ (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6))
+ async_submit = 0;
+ else
+ async_submit = 1;
+
bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
if (!bio)
return -ENOMEM;
@@ -6480,7 +7208,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
bio->bi_end_io = btrfs_end_dio_bio;
map_length = orig_bio->bi_size;
- ret = btrfs_map_block(root->fs_info, READ,
+ ret = btrfs_map_block(root->fs_info, rw,
start_sector << 9,
&map_length, NULL, 0);
if (ret) {
@@ -6623,15 +7351,60 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
+ size_t count = 0;
+ int flags = 0;
+ bool wakeup = true;
+ bool relock = false;
+ ssize_t ret;
if (check_direct_IO(BTRFS_I(inode)->root, rw, iocb, iov,
offset, nr_segs))
return 0;
- return __blockdev_direct_IO(rw, iocb, inode,
- BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev,
- iov, offset, nr_segs, btrfs_get_blocks_direct, NULL,
- btrfs_submit_direct, 0);
+ atomic_inc(&inode->i_dio_count);
+ smp_mb__after_atomic_inc();
+
+ if (rw & WRITE) {
+ count = iov_length(iov, nr_segs);
+ /*
+ * If the write DIO is beyond the EOF, we need update
+ * the isize, but it is protected by i_mutex. So we can
+ * not unlock the i_mutex at this case.
+ */
+ if (offset + count <= inode->i_size) {
+ mutex_unlock(&inode->i_mutex);
+ relock = true;
+ }
+ ret = btrfs_delalloc_reserve_space(inode, count);
+ if (ret)
+ goto out;
+ } else if (unlikely(test_bit(BTRFS_INODE_READDIO_NEED_LOCK,
+ &BTRFS_I(inode)->runtime_flags))) {
+ inode_dio_done(inode);
+ flags = DIO_LOCKING | DIO_SKIP_HOLES;
+ wakeup = false;
+ }
+
+ ret = __blockdev_direct_IO(rw, iocb, inode,
+ BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev,
+ iov, offset, nr_segs, btrfs_get_blocks_direct, NULL,
+ btrfs_submit_direct, flags);
+ if (rw & WRITE) {
+ if (ret < 0 && ret != -EIOCBQUEUED)
+ btrfs_delalloc_release_space(inode, count);
+ else if (ret >= 0 && (size_t)ret < count)
+ btrfs_delalloc_release_space(inode,
+ count - (size_t)ret);
+ else
+ btrfs_delalloc_release_metadata(inode, 0);
+ }
+out:
+ if (wakeup)
+ inode_dio_done(inode);
+ if (relock)
+ mutex_lock(&inode->i_mutex);
+
+ return ret;
}
#define BTRFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC)
@@ -6669,8 +7442,8 @@ static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
return extent_write_full_page(tree, page, btrfs_get_extent, wbc);
}
-int btrfs_writepages(struct address_space *mapping,
- struct writeback_control *wbc)
+static int btrfs_writepages(struct address_space *mapping,
+ struct writeback_control *wbc)
{
struct extent_io_tree *tree;
@@ -6735,8 +7508,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
return;
}
lock_extent_bits(tree, page_start, page_end, 0, &cached_state);
- ordered = btrfs_lookup_ordered_extent(inode,
- page_offset(page));
+ ordered = btrfs_lookup_ordered_extent(inode, page_offset(page));
if (ordered) {
/*
* IO on this page will never be started, so we need
@@ -6791,7 +7563,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
struct page *page = vmf->page;
- struct inode *inode = fdentry(vma->vm_file)->d_inode;
+ struct inode *inode = file_inode(vma->vm_file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
struct btrfs_ordered_extent *ordered;
@@ -7186,8 +7958,8 @@ void btrfs_destroy_inode(struct inode *inode)
if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
&BTRFS_I(inode)->runtime_flags)) {
- printk(KERN_INFO "BTRFS: inode %llu still on the orphan list\n",
- (unsigned long long)btrfs_ino(inode));
+ btrfs_info(root->fs_info, "inode %llu still on the orphan list",
+ (unsigned long long)btrfs_ino(inode));
atomic_dec(&root->orphan_inodes);
}
@@ -7196,10 +7968,9 @@ void btrfs_destroy_inode(struct inode *inode)
if (!ordered)
break;
else {
- printk(KERN_ERR "btrfs found ordered "
- "extent %llu %llu on inode cleanup\n",
- (unsigned long long)ordered->file_offset,
- (unsigned long long)ordered->len);
+ btrfs_err(root->fs_info, "found ordered extent %llu %llu on inode cleanup",
+ (unsigned long long)ordered->file_offset,
+ (unsigned long long)ordered->len);
btrfs_remove_ordered_extent(inode, ordered);
btrfs_put_ordered_extent(ordered);
btrfs_put_ordered_extent(ordered);
@@ -7216,8 +7987,9 @@ int btrfs_drop_inode(struct inode *inode)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
+ /* the snap/subvol tree is on deleting */
if (btrfs_root_refs(&root->root_item) == 0 &&
- !btrfs_is_free_space_inode(inode))
+ root != root->fs_info->tree_root)
return 1;
else
return generic_drop_inode(inode);
@@ -7299,40 +8071,22 @@ fail:
static int btrfs_getattr(struct vfsmount *mnt,
struct dentry *dentry, struct kstat *stat)
{
+ u64 delalloc_bytes;
struct inode *inode = dentry->d_inode;
u32 blocksize = inode->i_sb->s_blocksize;
generic_fillattr(inode, stat);
stat->dev = BTRFS_I(inode)->root->anon_dev;
stat->blksize = PAGE_CACHE_SIZE;
+
+ spin_lock(&BTRFS_I(inode)->lock);
+ delalloc_bytes = BTRFS_I(inode)->delalloc_bytes;
+ spin_unlock(&BTRFS_I(inode)->lock);
stat->blocks = (ALIGN(inode_get_bytes(inode), blocksize) +
- ALIGN(BTRFS_I(inode)->delalloc_bytes, blocksize)) >> 9;
+ ALIGN(delalloc_bytes, blocksize)) >> 9;
return 0;
}
-/*
- * If a file is moved, it will inherit the cow and compression flags of the new
- * directory.
- */
-static void fixup_inode_flags(struct inode *dir, struct inode *inode)
-{
- struct btrfs_inode *b_dir = BTRFS_I(dir);
- struct btrfs_inode *b_inode = BTRFS_I(inode);
-
- if (b_dir->flags & BTRFS_INODE_NODATACOW)
- b_inode->flags |= BTRFS_INODE_NODATACOW;
- else
- b_inode->flags &= ~BTRFS_INODE_NODATACOW;
-
- if (b_dir->flags & BTRFS_INODE_COMPRESS) {
- b_inode->flags |= BTRFS_INODE_COMPRESS;
- b_inode->flags &= ~BTRFS_INODE_NOCOMPRESS;
- } else {
- b_inode->flags &= ~(BTRFS_INODE_COMPRESS |
- BTRFS_INODE_NOCOMPRESS);
- }
-}
-
static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
struct inode *new_dir, struct dentry *new_dentry)
{
@@ -7403,7 +8157,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
* inodes. So 5 * 2 is 10, plus 1 for the new link, so 11 total items
* should cover the worst case number of items we'll modify.
*/
- trans = btrfs_start_transaction(root, 20);
+ trans = btrfs_start_transaction(root, 11);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto out_notrans;
@@ -7498,8 +8252,6 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
}
}
- fixup_inode_flags(new_dir, old_inode);
-
ret = btrfs_add_link(trans, new_dir, old_inode,
new_dentry->d_name.name,
new_dentry->d_name.len, 0, index);
@@ -7583,7 +8335,7 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
INIT_LIST_HEAD(&works);
INIT_LIST_HEAD(&splice);
-again:
+
spin_lock(&root->fs_info->delalloc_lock);
list_splice_init(&root->fs_info->delalloc_inodes, &splice);
while (!list_empty(&splice)) {
@@ -7593,8 +8345,11 @@ again:
list_del_init(&binode->delalloc_inodes);
inode = igrab(&binode->vfs_inode);
- if (!inode)
+ if (!inode) {
+ clear_bit(BTRFS_INODE_IN_DELALLOC_LIST,
+ &binode->runtime_flags);
continue;
+ }
list_add_tail(&binode->delalloc_inodes,
&root->fs_info->delalloc_inodes);
@@ -7619,13 +8374,6 @@ again:
btrfs_wait_and_free_delalloc_work(work);
}
- spin_lock(&root->fs_info->delalloc_lock);
- if (!list_empty(&root->fs_info->delalloc_inodes)) {
- spin_unlock(&root->fs_info->delalloc_lock);
- goto again;
- }
- spin_unlock(&root->fs_info->delalloc_lock);
-
/* the filemap_flush will queue IO into the worker threads, but
* we have to make sure the IO is actually started and that
* ordered extents get created before we return
@@ -7787,6 +8535,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
struct btrfs_key ins;
u64 cur_offset = start;
u64 i_size;
+ u64 cur_bytes;
int ret = 0;
bool own_trans = true;
@@ -7801,8 +8550,10 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
}
}
- ret = btrfs_reserve_extent(trans, root, num_bytes, min_size,
- 0, *alloc_hint, &ins, 1);
+ cur_bytes = min(num_bytes, 256ULL * 1024 * 1024);
+ cur_bytes = max(cur_bytes, min_size);
+ ret = btrfs_reserve_extent(trans, root, cur_bytes,
+ min_size, 0, *alloc_hint, &ins, 1);
if (ret) {
if (own_trans)
btrfs_end_transaction(trans, root);
@@ -7836,16 +8587,14 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
em->block_start = ins.objectid;
em->block_len = ins.offset;
em->orig_block_len = ins.offset;
+ em->ram_bytes = ins.offset;
em->bdev = root->fs_info->fs_devices->latest_bdev;
set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
em->generation = trans->transid;
while (1) {
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- if (!ret)
- list_move(&em->list,
- &em_tree->modified_extents);
+ ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
if (ret != -EEXIST)
break;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 338f2597bf7..0de4a2fcfb2 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -42,12 +42,12 @@
#include <linux/slab.h>
#include <linux/blkdev.h>
#include <linux/uuid.h>
+#include <linux/btrfs.h>
#include "compat.h"
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
-#include "ioctl.h"
#include "print-tree.h"
#include "volumes.h"
#include "locking.h"
@@ -152,7 +152,7 @@ void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
static int btrfs_ioctl_getflags(struct file *file, void __user *arg)
{
- struct btrfs_inode *ip = BTRFS_I(file->f_path.dentry->d_inode);
+ struct btrfs_inode *ip = BTRFS_I(file_inode(file));
unsigned int flags = btrfs_flags_to_ioctl(ip->flags);
if (copy_to_user(arg, &flags, sizeof(flags)))
@@ -177,7 +177,7 @@ static int check_flags(unsigned int flags)
static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
{
- struct inode *inode = file->f_path.dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_inode *ip = BTRFS_I(inode);
struct btrfs_root *root = ip->root;
struct btrfs_trans_handle *trans;
@@ -310,7 +310,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
static int btrfs_ioctl_getversion(struct file *file, int __user *arg)
{
- struct inode *inode = file->f_path.dentry->d_inode;
+ struct inode *inode = file_inode(file);
return put_user(inode->i_generation, arg);
}
@@ -363,46 +363,52 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
return 0;
}
-static noinline int create_subvol(struct btrfs_root *root,
+static noinline int create_subvol(struct inode *dir,
struct dentry *dentry,
char *name, int namelen,
u64 *async_transid,
- struct btrfs_qgroup_inherit **inherit)
+ struct btrfs_qgroup_inherit *inherit)
{
struct btrfs_trans_handle *trans;
struct btrfs_key key;
struct btrfs_root_item root_item;
struct btrfs_inode_item *inode_item;
struct extent_buffer *leaf;
+ struct btrfs_root *root = BTRFS_I(dir)->root;
struct btrfs_root *new_root;
- struct dentry *parent = dentry->d_parent;
- struct inode *dir;
+ struct btrfs_block_rsv block_rsv;
struct timespec cur_time = CURRENT_TIME;
int ret;
int err;
u64 objectid;
u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
u64 index = 0;
+ u64 qgroup_reserved;
uuid_le new_uuid;
ret = btrfs_find_free_objectid(root->fs_info->tree_root, &objectid);
if (ret)
return ret;
- dir = parent->d_inode;
-
+ btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
/*
- * 1 - inode item
- * 2 - refs
- * 1 - root item
- * 2 - dir items
+ * The same as the snapshot creation, please see the comment
+ * of create_snapshot().
*/
- trans = btrfs_start_transaction(root, 6);
- if (IS_ERR(trans))
- return PTR_ERR(trans);
+ ret = btrfs_subvolume_reserve_metadata(root, &block_rsv,
+ 7, &qgroup_reserved);
+ if (ret)
+ return ret;
+
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+ trans->block_rsv = &block_rsv;
+ trans->bytes_reserved = block_rsv.size;
- ret = btrfs_qgroup_inherit(trans, root->fs_info, 0, objectid,
- inherit ? *inherit : NULL);
+ ret = btrfs_qgroup_inherit(trans, root->fs_info, 0, objectid, inherit);
if (ret)
goto fail;
@@ -516,9 +522,13 @@ static noinline int create_subvol(struct btrfs_root *root,
BUG_ON(ret);
fail:
+ trans->block_rsv = NULL;
+ trans->bytes_reserved = 0;
if (async_transid) {
*async_transid = trans->transid;
err = btrfs_commit_transaction_async(trans, root, 1);
+ if (err)
+ err = btrfs_commit_transaction(trans, root);
} else {
err = btrfs_commit_transaction(trans, root);
}
@@ -527,13 +537,15 @@ fail:
if (!ret)
d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
-
+out:
+ btrfs_subvolume_release_metadata(root, &block_rsv, qgroup_reserved);
return ret;
}
-static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
- char *name, int namelen, u64 *async_transid,
- bool readonly, struct btrfs_qgroup_inherit **inherit)
+static int create_snapshot(struct btrfs_root *root, struct inode *dir,
+ struct dentry *dentry, char *name, int namelen,
+ u64 *async_transid, bool readonly,
+ struct btrfs_qgroup_inherit *inherit)
{
struct inode *inode;
struct btrfs_pending_snapshot *pending_snapshot;
@@ -549,23 +561,31 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
btrfs_init_block_rsv(&pending_snapshot->block_rsv,
BTRFS_BLOCK_RSV_TEMP);
+ /*
+ * 1 - parent dir inode
+ * 2 - dir entries
+ * 1 - root item
+ * 2 - root ref/backref
+ * 1 - root of snapshot
+ */
+ ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root,
+ &pending_snapshot->block_rsv, 7,
+ &pending_snapshot->qgroup_reserved);
+ if (ret)
+ goto out;
+
pending_snapshot->dentry = dentry;
pending_snapshot->root = root;
pending_snapshot->readonly = readonly;
- if (inherit) {
- pending_snapshot->inherit = *inherit;
- *inherit = NULL; /* take responsibility to free it */
- }
+ pending_snapshot->dir = dir;
+ pending_snapshot->inherit = inherit;
- trans = btrfs_start_transaction(root->fs_info->extent_root, 6);
+ trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto fail;
}
- ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
- BUG_ON(ret);
-
spin_lock(&root->fs_info->trans_lock);
list_add(&pending_snapshot->list,
&trans->transaction->pending_snapshots);
@@ -574,16 +594,14 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
*async_transid = trans->transid;
ret = btrfs_commit_transaction_async(trans,
root->fs_info->extent_root, 1);
+ if (ret)
+ ret = btrfs_commit_transaction(trans, root);
} else {
ret = btrfs_commit_transaction(trans,
root->fs_info->extent_root);
}
- if (ret) {
- /* cleanup_transaction has freed this for us */
- if (trans->aborted)
- pending_snapshot = NULL;
+ if (ret)
goto fail;
- }
ret = pending_snapshot->error;
if (ret)
@@ -602,6 +620,10 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
d_instantiate(dentry, inode);
ret = 0;
fail:
+ btrfs_subvolume_release_metadata(BTRFS_I(dir)->root,
+ &pending_snapshot->block_rsv,
+ pending_snapshot->qgroup_reserved);
+out:
kfree(pending_snapshot);
return ret;
}
@@ -695,13 +717,15 @@ static noinline int btrfs_mksubvol(struct path *parent,
char *name, int namelen,
struct btrfs_root *snap_src,
u64 *async_transid, bool readonly,
- struct btrfs_qgroup_inherit **inherit)
+ struct btrfs_qgroup_inherit *inherit)
{
struct inode *dir = parent->dentry->d_inode;
struct dentry *dentry;
int error;
- mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
+ error = mutex_lock_killable_nested(&dir->i_mutex, I_MUTEX_PARENT);
+ if (error == -EINTR)
+ return error;
dentry = lookup_one_len(name, parent->dentry, namelen);
error = PTR_ERR(dentry);
@@ -732,11 +756,11 @@ static noinline int btrfs_mksubvol(struct path *parent,
goto out_up_read;
if (snap_src) {
- error = create_snapshot(snap_src, dentry, name, namelen,
+ error = create_snapshot(snap_src, dir, dentry, name, namelen,
async_transid, readonly, inherit);
} else {
- error = create_subvol(BTRFS_I(dir)->root, dentry,
- name, namelen, async_transid, inherit);
+ error = create_subvol(dir, dentry, name, namelen,
+ async_transid, inherit);
}
if (!error)
fsnotify_mkdir(dir, dentry);
@@ -818,7 +842,7 @@ static int find_new_extents(struct btrfs_root *root,
while(1) {
ret = btrfs_search_forward(root, &min_key, &max_key,
- path, 0, newer_than);
+ path, newer_than);
if (ret != 0)
goto none;
if (min_key.objectid != ino)
@@ -1130,8 +1154,11 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
u64 new_align = ~((u64)128 * 1024 - 1);
struct page **pages = NULL;
- if (extent_thresh == 0)
- extent_thresh = 256 * 1024;
+ if (isize == 0)
+ return 0;
+
+ if (range->start >= isize)
+ return -EINVAL;
if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS) {
if (range->compress_type > BTRFS_COMPRESS_TYPES)
@@ -1140,8 +1167,8 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
compress_type = range->compress_type;
}
- if (isize == 0)
- return 0;
+ if (extent_thresh == 0)
+ extent_thresh = 256 * 1024;
/*
* if we were not given a file, allocate a readahead
@@ -1206,6 +1233,12 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
if (!(inode->i_sb->s_flags & MS_ACTIVE))
break;
+ if (btrfs_defrag_cancelled(root->fs_info)) {
+ printk(KERN_DEBUG "btrfs: defrag_file cancelled\n");
+ ret = -EAGAIN;
+ break;
+ }
+
if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT,
extent_thresh, &last_len, &skip,
&defrag_end, range->flags &
@@ -1320,7 +1353,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
u64 new_size;
u64 old_size;
u64 devid = 1;
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_vol_args *vol_args;
struct btrfs_trans_handle *trans;
struct btrfs_device *device = NULL;
@@ -1329,9 +1362,6 @@ static noinline int btrfs_ioctl_resize(struct file *file,
int ret = 0;
int mod = 0;
- if (root->fs_info->sb->s_flags & MS_RDONLY)
- return -EROFS;
-
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
@@ -1363,6 +1393,10 @@ static noinline int btrfs_ioctl_resize(struct file *file,
*devstr = '\0';
devstr = vol_args->name;
devid = simple_strtoull(devstr, &end, 10);
+ if (!devid) {
+ ret = -EINVAL;
+ goto out_free;
+ }
printk(KERN_INFO "btrfs: resizing devid %llu\n",
(unsigned long long)devid);
}
@@ -1371,7 +1405,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
if (!device) {
printk(KERN_INFO "btrfs: resizer unable to find device %llu\n",
(unsigned long long)devid);
- ret = -EINVAL;
+ ret = -ENODEV;
goto out_free;
}
@@ -1379,7 +1413,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
printk(KERN_INFO "btrfs: resizer unable to apply on "
"readonly device %llu\n",
(unsigned long long)devid);
- ret = -EINVAL;
+ ret = -EPERM;
goto out_free;
}
@@ -1401,7 +1435,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
}
if (device->is_tgtdev_for_dev_replace) {
- ret = -EINVAL;
+ ret = -EPERM;
goto out_free;
}
@@ -1457,7 +1491,7 @@ out:
static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
char *name, unsigned long fd, int subvol,
u64 *transid, bool readonly,
- struct btrfs_qgroup_inherit **inherit)
+ struct btrfs_qgroup_inherit *inherit)
{
int namelen;
int ret = 0;
@@ -1489,8 +1523,8 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
goto out_drop_write;
}
- src_inode = src.file->f_path.dentry->d_inode;
- if (src_inode->i_sb != file->f_path.dentry->d_inode->i_sb) {
+ src_inode = file_inode(src.file);
+ if (src_inode->i_sb != file_inode(file)->i_sb) {
printk(KERN_INFO "btrfs: Snapshot src from "
"another FS\n");
ret = -EINVAL;
@@ -1566,7 +1600,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
vol_args->fd, subvol, ptr,
- readonly, &inherit);
+ readonly, inherit);
if (ret == 0 && ptr &&
copy_to_user(arg +
@@ -1582,7 +1616,7 @@ out:
static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
void __user *arg)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret = 0;
u64 flags = 0;
@@ -1604,7 +1638,7 @@ static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
void __user *arg)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
u64 root_flags;
@@ -1863,7 +1897,7 @@ static noinline int search_ioctl(struct inode *inode,
path->keep_locks = 1;
while(1) {
- ret = btrfs_search_forward(root, &key, &max_key, path, 0,
+ ret = btrfs_search_forward(root, &key, &max_key, path,
sk->min_transid);
if (ret != 0) {
if (ret > 0)
@@ -1898,7 +1932,7 @@ static noinline int btrfs_ioctl_tree_search(struct file *file,
if (IS_ERR(args))
return PTR_ERR(args);
- inode = fdentry(file)->d_inode;
+ inode = file_inode(file);
ret = search_ioctl(inode, args);
if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
ret = -EFAULT;
@@ -2008,7 +2042,7 @@ static noinline int btrfs_ioctl_ino_lookup(struct file *file,
if (IS_ERR(args))
return PTR_ERR(args);
- inode = fdentry(file)->d_inode;
+ inode = file_inode(file);
if (args->treeid == 0)
args->treeid = BTRFS_I(inode)->root->root_key.objectid;
@@ -2035,6 +2069,8 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
struct btrfs_root *dest = NULL;
struct btrfs_ioctl_vol_args *vol_args;
struct btrfs_trans_handle *trans;
+ struct btrfs_block_rsv block_rsv;
+ u64 qgroup_reserved;
int namelen;
int ret;
int err = 0;
@@ -2055,7 +2091,9 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
if (err)
goto out;
- mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
+ err = mutex_lock_killable_nested(&dir->i_mutex, I_MUTEX_PARENT);
+ if (err == -EINTR)
+ goto out;
dentry = lookup_one_len(vol_args->name, parent, namelen);
if (IS_ERR(dentry)) {
err = PTR_ERR(dentry);
@@ -2124,12 +2162,23 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
if (err)
goto out_up_write;
+ btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
+ /*
+ * One for dir inode, two for dir entries, two for root
+ * ref/backref.
+ */
+ err = btrfs_subvolume_reserve_metadata(root, &block_rsv,
+ 5, &qgroup_reserved);
+ if (err)
+ goto out_up_write;
+
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
- goto out_up_write;
+ goto out_release;
}
- trans->block_rsv = &root->fs_info->global_block_rsv;
+ trans->block_rsv = &block_rsv;
+ trans->bytes_reserved = block_rsv.size;
ret = btrfs_unlink_subvol(trans, root, dir,
dest->root_key.objectid,
@@ -2159,10 +2208,14 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
}
}
out_end_trans:
+ trans->block_rsv = NULL;
+ trans->bytes_reserved = 0;
ret = btrfs_end_transaction(trans, root);
if (ret && !err)
err = ret;
inode->i_flags |= S_DEAD;
+out_release:
+ btrfs_subvolume_release_metadata(root, &block_rsv, qgroup_reserved);
out_up_write:
up_write(&root->fs_info->subvol_sem);
out_unlock:
@@ -2171,6 +2224,12 @@ out_unlock:
shrink_dcache_sb(root->fs_info->sb);
btrfs_invalidate_inodes(dest);
d_delete(dentry);
+
+ /* the last ref */
+ if (dest->cache_inode) {
+ iput(dest->cache_inode);
+ dest->cache_inode = NULL;
+ }
}
out_dput:
dput(dentry);
@@ -2184,7 +2243,7 @@ out:
static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_ioctl_defrag_range_args *range;
int ret;
@@ -2193,13 +2252,6 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
if (ret)
return ret;
- if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
- 1)) {
- pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
- mnt_drop_write_file(file);
- return -EINVAL;
- }
-
if (btrfs_root_readonly(root)) {
ret = -EROFS;
goto out;
@@ -2211,10 +2263,10 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
ret = -EPERM;
goto out;
}
- ret = btrfs_defrag_root(root, 0);
+ ret = btrfs_defrag_root(root);
if (ret)
goto out;
- ret = btrfs_defrag_root(root->fs_info->extent_root, 0);
+ ret = btrfs_defrag_root(root->fs_info->extent_root);
break;
case S_IFREG:
if (!(file->f_mode & FMODE_WRITE)) {
@@ -2244,7 +2296,7 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
/* the rest are all set to zero by kzalloc */
range->len = (u64)-1;
}
- ret = btrfs_defrag_file(fdentry(file)->d_inode, file,
+ ret = btrfs_defrag_file(file_inode(file), file,
range, 0, 0);
if (ret > 0)
ret = 0;
@@ -2254,7 +2306,6 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
ret = -EINVAL;
}
out:
- atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
mnt_drop_write_file(file);
return ret;
}
@@ -2292,7 +2343,7 @@ out:
static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_vol_args *vol_args;
int ret;
@@ -2381,7 +2432,6 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg)
mutex_lock(&fs_devices->device_list_mutex);
dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, NULL);
- mutex_unlock(&fs_devices->device_list_mutex);
if (!dev) {
ret = -ENODEV;
@@ -2405,6 +2455,7 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg)
}
out:
+ mutex_unlock(&fs_devices->device_list_mutex);
if (ret == 0 && copy_to_user(arg, di_args, sizeof(*di_args)))
ret = -EFAULT;
@@ -2415,7 +2466,7 @@ out:
static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
u64 off, u64 olen, u64 destoff)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct fd src_file;
struct inode *src;
@@ -2461,7 +2512,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
if (src_file.file->f_path.mnt != file->f_path.mnt)
goto out_fput;
- src = src_file.file->f_dentry->d_inode;
+ src = file_inode(src_file.file);
ret = -EINVAL;
if (src == inode)
@@ -2823,7 +2874,7 @@ static long btrfs_ioctl_clone_range(struct file *file, void __user *argp)
*/
static long btrfs_ioctl_trans_start(struct file *file)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
int ret;
@@ -2863,7 +2914,7 @@ out:
static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_root *new_root;
struct btrfs_dir_item *di;
@@ -2959,7 +3010,7 @@ void btrfs_get_block_group_info(struct list_head *groups_list,
}
}
-long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
+static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
{
struct btrfs_ioctl_space_args space_args;
struct btrfs_ioctl_space_info space;
@@ -3087,7 +3138,7 @@ out:
*/
long btrfs_ioctl_trans_end(struct file *file)
{
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
@@ -3111,7 +3162,7 @@ static noinline long btrfs_ioctl_start_sync(struct btrfs_root *root,
u64 transid;
int ret;
- trans = btrfs_attach_transaction(root);
+ trans = btrfs_attach_transaction_barrier(root);
if (IS_ERR(trans)) {
if (PTR_ERR(trans) != -ENOENT)
return PTR_ERR(trans);
@@ -3149,7 +3200,7 @@ static noinline long btrfs_ioctl_wait_sync(struct btrfs_root *root,
static long btrfs_ioctl_scrub(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_scrub_args *sa;
int ret;
@@ -3289,7 +3340,7 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg)
struct inode_fs_paths *ipath = NULL;
struct btrfs_path *path;
- if (!capable(CAP_SYS_ADMIN))
+ if (!capable(CAP_DAC_READ_SEARCH))
return -EPERM;
path = btrfs_alloc_path();
@@ -3440,7 +3491,7 @@ void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
static long btrfs_ioctl_balance(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_ioctl_balance_args *bargs;
struct btrfs_balance_control *bctl;
@@ -3630,7 +3681,7 @@ out:
static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_quota_ctl_args *sa;
struct btrfs_trans_handle *trans = NULL;
int ret;
@@ -3649,12 +3700,11 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
goto drop_write;
}
- if (sa->cmd != BTRFS_QUOTA_CTL_RESCAN) {
- trans = btrfs_start_transaction(root, 2);
- if (IS_ERR(trans)) {
- ret = PTR_ERR(trans);
- goto out;
- }
+ down_write(&root->fs_info->subvol_sem);
+ trans = btrfs_start_transaction(root->fs_info->tree_root, 2);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
}
switch (sa->cmd) {
@@ -3664,9 +3714,6 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
case BTRFS_QUOTA_CTL_DISABLE:
ret = btrfs_quota_disable(trans, root->fs_info);
break;
- case BTRFS_QUOTA_CTL_RESCAN:
- ret = btrfs_quota_rescan(root->fs_info);
- break;
default:
ret = -EINVAL;
break;
@@ -3675,13 +3722,12 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg)
if (copy_to_user(arg, sa, sizeof(*sa)))
ret = -EFAULT;
- if (trans) {
- err = btrfs_commit_transaction(trans, root);
- if (err && !ret)
- ret = err;
- }
+ err = btrfs_commit_transaction(trans, root->fs_info->tree_root);
+ if (err && !ret)
+ ret = err;
out:
kfree(sa);
+ up_write(&root->fs_info->subvol_sem);
drop_write:
mnt_drop_write_file(file);
return ret;
@@ -3689,7 +3735,7 @@ drop_write:
static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_qgroup_assign_args *sa;
struct btrfs_trans_handle *trans;
int ret;
@@ -3736,7 +3782,7 @@ drop_write:
static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_qgroup_create_args *sa;
struct btrfs_trans_handle *trans;
int ret;
@@ -3787,7 +3833,7 @@ drop_write:
static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
struct btrfs_ioctl_qgroup_limit_args *sa;
struct btrfs_trans_handle *trans;
int ret;
@@ -3833,11 +3879,69 @@ drop_write:
return ret;
}
+static long btrfs_ioctl_quota_rescan(struct file *file, void __user *arg)
+{
+ struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_ioctl_quota_rescan_args *qsa;
+ int ret;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ ret = mnt_want_write_file(file);
+ if (ret)
+ return ret;
+
+ qsa = memdup_user(arg, sizeof(*qsa));
+ if (IS_ERR(qsa)) {
+ ret = PTR_ERR(qsa);
+ goto drop_write;
+ }
+
+ if (qsa->flags) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ ret = btrfs_qgroup_rescan(root->fs_info);
+
+out:
+ kfree(qsa);
+drop_write:
+ mnt_drop_write_file(file);
+ return ret;
+}
+
+static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg)
+{
+ struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_ioctl_quota_rescan_args *qsa;
+ int ret = 0;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ qsa = kzalloc(sizeof(*qsa), GFP_NOFS);
+ if (!qsa)
+ return -ENOMEM;
+
+ if (root->fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+ qsa->flags = 1;
+ qsa->progress = root->fs_info->qgroup_rescan_progress.objectid;
+ }
+
+ if (copy_to_user(arg, qsa, sizeof(*qsa)))
+ ret = -EFAULT;
+
+ kfree(qsa);
+ return ret;
+}
+
static long btrfs_ioctl_set_received_subvol(struct file *file,
void __user *arg)
{
struct btrfs_ioctl_received_subvol_args *sa = NULL;
- struct inode *inode = fdentry(file)->d_inode;
+ struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_root_item *root_item = &root->root_item;
struct btrfs_trans_handle *trans;
@@ -3914,10 +4018,69 @@ out:
return ret;
}
+static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg)
+{
+ struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ const char *label = root->fs_info->super_copy->label;
+ size_t len = strnlen(label, BTRFS_LABEL_SIZE);
+ int ret;
+
+ if (len == BTRFS_LABEL_SIZE) {
+ pr_warn("btrfs: label is too long, return the first %zu bytes\n",
+ --len);
+ }
+
+ mutex_lock(&root->fs_info->volume_mutex);
+ ret = copy_to_user(arg, label, len);
+ mutex_unlock(&root->fs_info->volume_mutex);
+
+ return ret ? -EFAULT : 0;
+}
+
+static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg)
+{
+ struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_super_block *super_block = root->fs_info->super_copy;
+ struct btrfs_trans_handle *trans;
+ char label[BTRFS_LABEL_SIZE];
+ int ret;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ if (copy_from_user(label, arg, sizeof(label)))
+ return -EFAULT;
+
+ if (strnlen(label, BTRFS_LABEL_SIZE) == BTRFS_LABEL_SIZE) {
+ pr_err("btrfs: unable to set label with more than %d bytes\n",
+ BTRFS_LABEL_SIZE - 1);
+ return -EINVAL;
+ }
+
+ ret = mnt_want_write_file(file);
+ if (ret)
+ return ret;
+
+ mutex_lock(&root->fs_info->volume_mutex);
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out_unlock;
+ }
+
+ strcpy(super_block->label, label);
+ ret = btrfs_end_transaction(trans, root);
+
+out_unlock:
+ mutex_unlock(&root->fs_info->volume_mutex);
+ mnt_drop_write_file(file);
+ return ret;
+}
+
long btrfs_ioctl(struct file *file, unsigned int
cmd, unsigned long arg)
{
- struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+ struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
void __user *argp = (void __user *)arg;
switch (cmd) {
@@ -4012,8 +4175,16 @@ long btrfs_ioctl(struct file *file, unsigned int
return btrfs_ioctl_qgroup_create(file, argp);
case BTRFS_IOC_QGROUP_LIMIT:
return btrfs_ioctl_qgroup_limit(file, argp);
+ case BTRFS_IOC_QUOTA_RESCAN:
+ return btrfs_ioctl_quota_rescan(file, argp);
+ case BTRFS_IOC_QUOTA_RESCAN_STATUS:
+ return btrfs_ioctl_quota_rescan_status(file, argp);
case BTRFS_IOC_DEV_REPLACE:
return btrfs_ioctl_dev_replace(root, argp);
+ case BTRFS_IOC_GET_FSLABEL:
+ return btrfs_ioctl_get_fslabel(file, argp);
+ case BTRFS_IOC_SET_FSLABEL:
+ return btrfs_ioctl_set_fslabel(file, argp);
}
return -ENOTTY;
diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h
deleted file mode 100644
index dabca9cc8c2..00000000000
--- a/fs/btrfs/ioctl.h
+++ /dev/null
@@ -1,502 +0,0 @@
-/*
- * Copyright (C) 2007 Oracle. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public
- * License v2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public
- * License along with this program; if not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 021110-1307, USA.
- */
-
-#ifndef __IOCTL_
-#define __IOCTL_
-#include <linux/ioctl.h>
-
-#define BTRFS_IOCTL_MAGIC 0x94
-#define BTRFS_VOL_NAME_MAX 255
-
-/* this should be 4k */
-#define BTRFS_PATH_NAME_MAX 4087
-struct btrfs_ioctl_vol_args {
- __s64 fd;
- char name[BTRFS_PATH_NAME_MAX + 1];
-};
-
-#define BTRFS_DEVICE_PATH_NAME_MAX 1024
-
-#define BTRFS_SUBVOL_CREATE_ASYNC (1ULL << 0)
-#define BTRFS_SUBVOL_RDONLY (1ULL << 1)
-#define BTRFS_SUBVOL_QGROUP_INHERIT (1ULL << 2)
-#define BTRFS_FSID_SIZE 16
-#define BTRFS_UUID_SIZE 16
-
-#define BTRFS_QGROUP_INHERIT_SET_LIMITS (1ULL << 0)
-
-struct btrfs_qgroup_limit {
- __u64 flags;
- __u64 max_rfer;
- __u64 max_excl;
- __u64 rsv_rfer;
- __u64 rsv_excl;
-};
-
-struct btrfs_qgroup_inherit {
- __u64 flags;
- __u64 num_qgroups;
- __u64 num_ref_copies;
- __u64 num_excl_copies;
- struct btrfs_qgroup_limit lim;
- __u64 qgroups[0];
-};
-
-struct btrfs_ioctl_qgroup_limit_args {
- __u64 qgroupid;
- struct btrfs_qgroup_limit lim;
-};
-
-#define BTRFS_SUBVOL_NAME_MAX 4039
-struct btrfs_ioctl_vol_args_v2 {
- __s64 fd;
- __u64 transid;
- __u64 flags;
- union {
- struct {
- __u64 size;
- struct btrfs_qgroup_inherit __user *qgroup_inherit;
- };
- __u64 unused[4];
- };
- char name[BTRFS_SUBVOL_NAME_MAX + 1];
-};
-
-/*
- * structure to report errors and progress to userspace, either as a
- * result of a finished scrub, a canceled scrub or a progress inquiry
- */
-struct btrfs_scrub_progress {
- __u64 data_extents_scrubbed; /* # of data extents scrubbed */
- __u64 tree_extents_scrubbed; /* # of tree extents scrubbed */
- __u64 data_bytes_scrubbed; /* # of data bytes scrubbed */
- __u64 tree_bytes_scrubbed; /* # of tree bytes scrubbed */
- __u64 read_errors; /* # of read errors encountered (EIO) */
- __u64 csum_errors; /* # of failed csum checks */
- __u64 verify_errors; /* # of occurences, where the metadata
- * of a tree block did not match the
- * expected values, like generation or
- * logical */
- __u64 no_csum; /* # of 4k data block for which no csum
- * is present, probably the result of
- * data written with nodatasum */
- __u64 csum_discards; /* # of csum for which no data was found
- * in the extent tree. */
- __u64 super_errors; /* # of bad super blocks encountered */
- __u64 malloc_errors; /* # of internal kmalloc errors. These
- * will likely cause an incomplete
- * scrub */
- __u64 uncorrectable_errors; /* # of errors where either no intact
- * copy was found or the writeback
- * failed */
- __u64 corrected_errors; /* # of errors corrected */
- __u64 last_physical; /* last physical address scrubbed. In
- * case a scrub was aborted, this can
- * be used to restart the scrub */
- __u64 unverified_errors; /* # of occurences where a read for a
- * full (64k) bio failed, but the re-
- * check succeeded for each 4k piece.
- * Intermittent error. */
-};
-
-#define BTRFS_SCRUB_READONLY 1
-struct btrfs_ioctl_scrub_args {
- __u64 devid; /* in */
- __u64 start; /* in */
- __u64 end; /* in */
- __u64 flags; /* in */
- struct btrfs_scrub_progress progress; /* out */
- /* pad to 1k */
- __u64 unused[(1024-32-sizeof(struct btrfs_scrub_progress))/8];
-};
-
-#define BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS 0
-#define BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID 1
-struct btrfs_ioctl_dev_replace_start_params {
- __u64 srcdevid; /* in, if 0, use srcdev_name instead */
- __u64 cont_reading_from_srcdev_mode; /* in, see #define
- * above */
- __u8 srcdev_name[BTRFS_DEVICE_PATH_NAME_MAX + 1]; /* in */
- __u8 tgtdev_name[BTRFS_DEVICE_PATH_NAME_MAX + 1]; /* in */
-};
-
-#define BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED 0
-#define BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED 1
-#define BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED 2
-#define BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED 3
-#define BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED 4
-struct btrfs_ioctl_dev_replace_status_params {
- __u64 replace_state; /* out, see #define above */
- __u64 progress_1000; /* out, 0 <= x <= 1000 */
- __u64 time_started; /* out, seconds since 1-Jan-1970 */
- __u64 time_stopped; /* out, seconds since 1-Jan-1970 */
- __u64 num_write_errors; /* out */
- __u64 num_uncorrectable_read_errors; /* out */
-};
-
-#define BTRFS_IOCTL_DEV_REPLACE_CMD_START 0
-#define BTRFS_IOCTL_DEV_REPLACE_CMD_STATUS 1
-#define BTRFS_IOCTL_DEV_REPLACE_CMD_CANCEL 2
-#define BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR 0
-#define BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED 1
-#define BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED 2
-struct btrfs_ioctl_dev_replace_args {
- __u64 cmd; /* in */
- __u64 result; /* out */
-
- union {
- struct btrfs_ioctl_dev_replace_start_params start;
- struct btrfs_ioctl_dev_replace_status_params status;
- }; /* in/out */
-
- __u64 spare[64];
-};
-
-struct btrfs_ioctl_dev_info_args {
- __u64 devid; /* in/out */
- __u8 uuid[BTRFS_UUID_SIZE]; /* in/out */
- __u64 bytes_used; /* out */
- __u64 total_bytes; /* out */
- __u64 unused[379]; /* pad to 4k */
- __u8 path[BTRFS_DEVICE_PATH_NAME_MAX]; /* out */
-};
-
-struct btrfs_ioctl_fs_info_args {
- __u64 max_id; /* out */
- __u64 num_devices; /* out */
- __u8 fsid[BTRFS_FSID_SIZE]; /* out */
- __u64 reserved[124]; /* pad to 1k */
-};
-
-/* balance control ioctl modes */
-#define BTRFS_BALANCE_CTL_PAUSE 1
-#define BTRFS_BALANCE_CTL_CANCEL 2
-
-/*
- * this is packed, because it should be exactly the same as its disk
- * byte order counterpart (struct btrfs_disk_balance_args)
- */
-struct btrfs_balance_args {
- __u64 profiles;
- __u64 usage;
- __u64 devid;
- __u64 pstart;
- __u64 pend;
- __u64 vstart;
- __u64 vend;
-
- __u64 target;
-
- __u64 flags;
-
- __u64 unused[8];
-} __attribute__ ((__packed__));
-
-/* report balance progress to userspace */
-struct btrfs_balance_progress {
- __u64 expected; /* estimated # of chunks that will be
- * relocated to fulfill the request */
- __u64 considered; /* # of chunks we have considered so far */
- __u64 completed; /* # of chunks relocated so far */
-};
-
-#define BTRFS_BALANCE_STATE_RUNNING (1ULL << 0)
-#define BTRFS_BALANCE_STATE_PAUSE_REQ (1ULL << 1)
-#define BTRFS_BALANCE_STATE_CANCEL_REQ (1ULL << 2)
-
-struct btrfs_ioctl_balance_args {
- __u64 flags; /* in/out */
- __u64 state; /* out */
-
- struct btrfs_balance_args data; /* in/out */
- struct btrfs_balance_args meta; /* in/out */
- struct btrfs_balance_args sys; /* in/out */
-
- struct btrfs_balance_progress stat; /* out */
-
- __u64 unused[72]; /* pad to 1k */
-};
-
-#define BTRFS_INO_LOOKUP_PATH_MAX 4080
-struct btrfs_ioctl_ino_lookup_args {
- __u64 treeid;
- __u64 objectid;
- char name[BTRFS_INO_LOOKUP_PATH_MAX];
-};
-
-struct btrfs_ioctl_search_key {
- /* which root are we searching. 0 is the tree of tree roots */
- __u64 tree_id;
-
- /* keys returned will be >= min and <= max */
- __u64 min_objectid;
- __u64 max_objectid;
-
- /* keys returned will be >= min and <= max */
- __u64 min_offset;
- __u64 max_offset;
-
- /* max and min transids to search for */
- __u64 min_transid;
- __u64 max_transid;
-
- /* keys returned will be >= min and <= max */
- __u32 min_type;
- __u32 max_type;
-
- /*
- * how many items did userland ask for, and how many are we
- * returning
- */
- __u32 nr_items;
-
- /* align to 64 bits */
- __u32 unused;
-
- /* some extra for later */
- __u64 unused1;
- __u64 unused2;
- __u64 unused3;
- __u64 unused4;
-};
-
-struct btrfs_ioctl_search_header {
- __u64 transid;
- __u64 objectid;
- __u64 offset;
- __u32 type;
- __u32 len;
-};
-
-#define BTRFS_SEARCH_ARGS_BUFSIZE (4096 - sizeof(struct btrfs_ioctl_search_key))
-/*
- * the buf is an array of search headers where
- * each header is followed by the actual item
- * the type field is expanded to 32 bits for alignment
- */
-struct btrfs_ioctl_search_args {
- struct btrfs_ioctl_search_key key;
- char buf[BTRFS_SEARCH_ARGS_BUFSIZE];
-};
-
-struct btrfs_ioctl_clone_range_args {
- __s64 src_fd;
- __u64 src_offset, src_length;
- __u64 dest_offset;
-};
-
-/* flags for the defrag range ioctl */
-#define BTRFS_DEFRAG_RANGE_COMPRESS 1
-#define BTRFS_DEFRAG_RANGE_START_IO 2
-
-struct btrfs_ioctl_space_info {
- __u64 flags;
- __u64 total_bytes;
- __u64 used_bytes;
-};
-
-struct btrfs_ioctl_space_args {
- __u64 space_slots;
- __u64 total_spaces;
- struct btrfs_ioctl_space_info spaces[0];
-};
-
-struct btrfs_data_container {
- __u32 bytes_left; /* out -- bytes not needed to deliver output */
- __u32 bytes_missing; /* out -- additional bytes needed for result */
- __u32 elem_cnt; /* out */
- __u32 elem_missed; /* out */
- __u64 val[0]; /* out */
-};
-
-struct btrfs_ioctl_ino_path_args {
- __u64 inum; /* in */
- __u64 size; /* in */
- __u64 reserved[4];
- /* struct btrfs_data_container *fspath; out */
- __u64 fspath; /* out */
-};
-
-struct btrfs_ioctl_logical_ino_args {
- __u64 logical; /* in */
- __u64 size; /* in */
- __u64 reserved[4];
- /* struct btrfs_data_container *inodes; out */
- __u64 inodes;
-};
-
-enum btrfs_dev_stat_values {
- /* disk I/O failure stats */
- BTRFS_DEV_STAT_WRITE_ERRS, /* EIO or EREMOTEIO from lower layers */
- BTRFS_DEV_STAT_READ_ERRS, /* EIO or EREMOTEIO from lower layers */
- BTRFS_DEV_STAT_FLUSH_ERRS, /* EIO or EREMOTEIO from lower layers */
-
- /* stats for indirect indications for I/O failures */
- BTRFS_DEV_STAT_CORRUPTION_ERRS, /* checksum error, bytenr error or
- * contents is illegal: this is an
- * indication that the block was damaged
- * during read or write, or written to
- * wrong location or read from wrong
- * location */
- BTRFS_DEV_STAT_GENERATION_ERRS, /* an indication that blocks have not
- * been written */
-
- BTRFS_DEV_STAT_VALUES_MAX
-};
-
-/* Reset statistics after reading; needs SYS_ADMIN capability */
-#define BTRFS_DEV_STATS_RESET (1ULL << 0)
-
-struct btrfs_ioctl_get_dev_stats {
- __u64 devid; /* in */
- __u64 nr_items; /* in/out */
- __u64 flags; /* in/out */
-
- /* out values: */
- __u64 values[BTRFS_DEV_STAT_VALUES_MAX];
-
- __u64 unused[128 - 2 - BTRFS_DEV_STAT_VALUES_MAX]; /* pad to 1k */
-};
-
-#define BTRFS_QUOTA_CTL_ENABLE 1
-#define BTRFS_QUOTA_CTL_DISABLE 2
-#define BTRFS_QUOTA_CTL_RESCAN 3
-struct btrfs_ioctl_quota_ctl_args {
- __u64 cmd;
- __u64 status;
-};
-
-struct btrfs_ioctl_qgroup_assign_args {
- __u64 assign;
- __u64 src;
- __u64 dst;
-};
-
-struct btrfs_ioctl_qgroup_create_args {
- __u64 create;
- __u64 qgroupid;
-};
-struct btrfs_ioctl_timespec {
- __u64 sec;
- __u32 nsec;
-};
-
-struct btrfs_ioctl_received_subvol_args {
- char uuid[BTRFS_UUID_SIZE]; /* in */
- __u64 stransid; /* in */
- __u64 rtransid; /* out */
- struct btrfs_ioctl_timespec stime; /* in */
- struct btrfs_ioctl_timespec rtime; /* out */
- __u64 flags; /* in */
- __u64 reserved[16]; /* in */
-};
-
-struct btrfs_ioctl_send_args {
- __s64 send_fd; /* in */
- __u64 clone_sources_count; /* in */
- __u64 __user *clone_sources; /* in */
- __u64 parent_root; /* in */
- __u64 flags; /* in */
- __u64 reserved[4]; /* in */
-};
-
-#define BTRFS_IOC_SNAP_CREATE _IOW(BTRFS_IOCTL_MAGIC, 1, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_DEFRAG _IOW(BTRFS_IOCTL_MAGIC, 2, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_RESIZE _IOW(BTRFS_IOCTL_MAGIC, 3, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
- struct btrfs_ioctl_vol_args)
-/* trans start and trans end are dangerous, and only for
- * use by applications that know how to avoid the
- * resulting deadlocks
- */
-#define BTRFS_IOC_TRANS_START _IO(BTRFS_IOCTL_MAGIC, 6)
-#define BTRFS_IOC_TRANS_END _IO(BTRFS_IOCTL_MAGIC, 7)
-#define BTRFS_IOC_SYNC _IO(BTRFS_IOCTL_MAGIC, 8)
-
-#define BTRFS_IOC_CLONE _IOW(BTRFS_IOCTL_MAGIC, 9, int)
-#define BTRFS_IOC_ADD_DEV _IOW(BTRFS_IOCTL_MAGIC, 10, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_RM_DEV _IOW(BTRFS_IOCTL_MAGIC, 11, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_BALANCE _IOW(BTRFS_IOCTL_MAGIC, 12, \
- struct btrfs_ioctl_vol_args)
-
-#define BTRFS_IOC_CLONE_RANGE _IOW(BTRFS_IOCTL_MAGIC, 13, \
- struct btrfs_ioctl_clone_range_args)
-
-#define BTRFS_IOC_SUBVOL_CREATE _IOW(BTRFS_IOCTL_MAGIC, 14, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_SNAP_DESTROY _IOW(BTRFS_IOCTL_MAGIC, 15, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_DEFRAG_RANGE _IOW(BTRFS_IOCTL_MAGIC, 16, \
- struct btrfs_ioctl_defrag_range_args)
-#define BTRFS_IOC_TREE_SEARCH _IOWR(BTRFS_IOCTL_MAGIC, 17, \
- struct btrfs_ioctl_search_args)
-#define BTRFS_IOC_INO_LOOKUP _IOWR(BTRFS_IOCTL_MAGIC, 18, \
- struct btrfs_ioctl_ino_lookup_args)
-#define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, u64)
-#define BTRFS_IOC_SPACE_INFO _IOWR(BTRFS_IOCTL_MAGIC, 20, \
- struct btrfs_ioctl_space_args)
-#define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64)
-#define BTRFS_IOC_WAIT_SYNC _IOW(BTRFS_IOCTL_MAGIC, 22, __u64)
-#define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \
- struct btrfs_ioctl_vol_args_v2)
-#define BTRFS_IOC_SUBVOL_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 24, \
- struct btrfs_ioctl_vol_args_v2)
-#define BTRFS_IOC_SUBVOL_GETFLAGS _IOR(BTRFS_IOCTL_MAGIC, 25, __u64)
-#define BTRFS_IOC_SUBVOL_SETFLAGS _IOW(BTRFS_IOCTL_MAGIC, 26, __u64)
-#define BTRFS_IOC_SCRUB _IOWR(BTRFS_IOCTL_MAGIC, 27, \
- struct btrfs_ioctl_scrub_args)
-#define BTRFS_IOC_SCRUB_CANCEL _IO(BTRFS_IOCTL_MAGIC, 28)
-#define BTRFS_IOC_SCRUB_PROGRESS _IOWR(BTRFS_IOCTL_MAGIC, 29, \
- struct btrfs_ioctl_scrub_args)
-#define BTRFS_IOC_DEV_INFO _IOWR(BTRFS_IOCTL_MAGIC, 30, \
- struct btrfs_ioctl_dev_info_args)
-#define BTRFS_IOC_FS_INFO _IOR(BTRFS_IOCTL_MAGIC, 31, \
- struct btrfs_ioctl_fs_info_args)
-#define BTRFS_IOC_BALANCE_V2 _IOWR(BTRFS_IOCTL_MAGIC, 32, \
- struct btrfs_ioctl_balance_args)
-#define BTRFS_IOC_BALANCE_CTL _IOW(BTRFS_IOCTL_MAGIC, 33, int)
-#define BTRFS_IOC_BALANCE_PROGRESS _IOR(BTRFS_IOCTL_MAGIC, 34, \
- struct btrfs_ioctl_balance_args)
-#define BTRFS_IOC_INO_PATHS _IOWR(BTRFS_IOCTL_MAGIC, 35, \
- struct btrfs_ioctl_ino_path_args)
-#define BTRFS_IOC_LOGICAL_INO _IOWR(BTRFS_IOCTL_MAGIC, 36, \
- struct btrfs_ioctl_ino_path_args)
-#define BTRFS_IOC_SET_RECEIVED_SUBVOL _IOWR(BTRFS_IOCTL_MAGIC, 37, \
- struct btrfs_ioctl_received_subvol_args)
-#define BTRFS_IOC_SEND _IOW(BTRFS_IOCTL_MAGIC, 38, struct btrfs_ioctl_send_args)
-#define BTRFS_IOC_DEVICES_READY _IOR(BTRFS_IOCTL_MAGIC, 39, \
- struct btrfs_ioctl_vol_args)
-#define BTRFS_IOC_QUOTA_CTL _IOWR(BTRFS_IOCTL_MAGIC, 40, \
- struct btrfs_ioctl_quota_ctl_args)
-#define BTRFS_IOC_QGROUP_ASSIGN _IOW(BTRFS_IOCTL_MAGIC, 41, \
- struct btrfs_ioctl_qgroup_assign_args)
-#define BTRFS_IOC_QGROUP_CREATE _IOW(BTRFS_IOCTL_MAGIC, 42, \
- struct btrfs_ioctl_qgroup_create_args)
-#define BTRFS_IOC_QGROUP_LIMIT _IOR(BTRFS_IOCTL_MAGIC, 43, \
- struct btrfs_ioctl_qgroup_limit_args)
-#define BTRFS_IOC_GET_DEV_STATS _IOWR(BTRFS_IOCTL_MAGIC, 52, \
- struct btrfs_ioctl_get_dev_stats)
-#define BTRFS_IOC_DEV_REPLACE _IOWR(BTRFS_IOCTL_MAGIC, 53, \
- struct btrfs_ioctl_dev_replace_args)
-
-#endif
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index 2a1762c6604..01277b8f237 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -24,7 +24,7 @@
#include "extent_io.h"
#include "locking.h"
-void btrfs_assert_tree_read_locked(struct extent_buffer *eb);
+static void btrfs_assert_tree_read_locked(struct extent_buffer *eb);
/*
* if we currently have a spinning reader or writer lock
@@ -113,11 +113,10 @@ again:
read_unlock(&eb->lock);
return;
}
- read_unlock(&eb->lock);
- wait_event(eb->write_lock_wq, atomic_read(&eb->blocking_writers) == 0);
- read_lock(&eb->lock);
if (atomic_read(&eb->blocking_writers)) {
read_unlock(&eb->lock);
+ wait_event(eb->write_lock_wq,
+ atomic_read(&eb->blocking_writers) == 0);
goto again;
}
atomic_inc(&eb->read_locks);
@@ -265,7 +264,7 @@ void btrfs_assert_tree_locked(struct extent_buffer *eb)
BUG_ON(!atomic_read(&eb->write_locks));
}
-void btrfs_assert_tree_read_locked(struct extent_buffer *eb)
+static void btrfs_assert_tree_read_locked(struct extent_buffer *eb)
{
BUG_ON(!atomic_read(&eb->read_locks));
}
diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h
index ca52681e5f4..b81e0e9a489 100644
--- a/fs/btrfs/locking.h
+++ b/fs/btrfs/locking.h
@@ -26,7 +26,6 @@
void btrfs_tree_lock(struct extent_buffer *eb);
void btrfs_tree_unlock(struct extent_buffer *eb);
-int btrfs_try_spin_lock(struct extent_buffer *eb);
void btrfs_tree_read_lock(struct extent_buffer *eb);
void btrfs_tree_read_unlock(struct extent_buffer *eb);
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index e5ed5672960..1ddd728541e 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -196,6 +196,9 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
entry->file_offset = file_offset;
entry->start = start;
entry->len = len;
+ if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) &&
+ !(type == BTRFS_ORDERED_NOCOW))
+ entry->csum_bytes_left = disk_len;
entry->disk_len = disk_len;
entry->bytes_left = len;
entry->inode = igrab(inode);
@@ -213,6 +216,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
INIT_LIST_HEAD(&entry->root_extent_list);
INIT_LIST_HEAD(&entry->work_list);
init_completion(&entry->completion);
+ INIT_LIST_HEAD(&entry->log_list);
trace_btrfs_ordered_extent_add(inode, entry);
@@ -270,6 +274,10 @@ void btrfs_add_ordered_sum(struct inode *inode,
tree = &BTRFS_I(inode)->ordered_tree;
spin_lock_irq(&tree->lock);
list_add_tail(&sum->list, &entry->list);
+ WARN_ON(entry->csum_bytes_left < sum->len);
+ entry->csum_bytes_left -= sum->len;
+ if (entry->csum_bytes_left == 0)
+ wake_up(&entry->wait);
spin_unlock_irq(&tree->lock);
}
@@ -405,6 +413,66 @@ out:
return ret == 0;
}
+/* Needs to either be called under a log transaction or the log_mutex */
+void btrfs_get_logged_extents(struct btrfs_root *log, struct inode *inode)
+{
+ struct btrfs_ordered_inode_tree *tree;
+ struct btrfs_ordered_extent *ordered;
+ struct rb_node *n;
+ int index = log->log_transid % 2;
+
+ tree = &BTRFS_I(inode)->ordered_tree;
+ spin_lock_irq(&tree->lock);
+ for (n = rb_first(&tree->tree); n; n = rb_next(n)) {
+ ordered = rb_entry(n, struct btrfs_ordered_extent, rb_node);
+ spin_lock(&log->log_extents_lock[index]);
+ if (list_empty(&ordered->log_list)) {
+ list_add_tail(&ordered->log_list, &log->logged_list[index]);
+ atomic_inc(&ordered->refs);
+ }
+ spin_unlock(&log->log_extents_lock[index]);
+ }
+ spin_unlock_irq(&tree->lock);
+}
+
+void btrfs_wait_logged_extents(struct btrfs_root *log, u64 transid)
+{
+ struct btrfs_ordered_extent *ordered;
+ int index = transid % 2;
+
+ spin_lock_irq(&log->log_extents_lock[index]);
+ while (!list_empty(&log->logged_list[index])) {
+ ordered = list_first_entry(&log->logged_list[index],
+ struct btrfs_ordered_extent,
+ log_list);
+ list_del_init(&ordered->log_list);
+ spin_unlock_irq(&log->log_extents_lock[index]);
+ wait_event(ordered->wait, test_bit(BTRFS_ORDERED_IO_DONE,
+ &ordered->flags));
+ btrfs_put_ordered_extent(ordered);
+ spin_lock_irq(&log->log_extents_lock[index]);
+ }
+ spin_unlock_irq(&log->log_extents_lock[index]);
+}
+
+void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid)
+{
+ struct btrfs_ordered_extent *ordered;
+ int index = transid % 2;
+
+ spin_lock_irq(&log->log_extents_lock[index]);
+ while (!list_empty(&log->logged_list[index])) {
+ ordered = list_first_entry(&log->logged_list[index],
+ struct btrfs_ordered_extent,
+ log_list);
+ list_del_init(&ordered->log_list);
+ spin_unlock_irq(&log->log_extents_lock[index]);
+ btrfs_put_ordered_extent(ordered);
+ spin_lock_irq(&log->log_extents_lock[index]);
+ }
+ spin_unlock_irq(&log->log_extents_lock[index]);
+}
+
/*
* used to drop a reference on an ordered extent. This will free
* the extent if the last reference is dropped
@@ -489,6 +557,7 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput)
INIT_LIST_HEAD(&splice);
INIT_LIST_HEAD(&works);
+ mutex_lock(&root->fs_info->ordered_operations_mutex);
spin_lock(&root->fs_info->ordered_extent_lock);
list_splice_init(&root->fs_info->ordered_extents, &splice);
while (!list_empty(&splice)) {
@@ -532,6 +601,7 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput)
cond_resched();
}
+ mutex_unlock(&root->fs_info->ordered_operations_mutex);
}
/*
@@ -544,10 +614,12 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput)
* extra check to make sure the ordered operation list really is empty
* before we return
*/
-int btrfs_run_ordered_operations(struct btrfs_root *root, int wait)
+int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, int wait)
{
struct btrfs_inode *btrfs_inode;
struct inode *inode;
+ struct btrfs_transaction *cur_trans = trans->transaction;
struct list_head splice;
struct list_head works;
struct btrfs_delalloc_work *work, *next;
@@ -558,14 +630,10 @@ int btrfs_run_ordered_operations(struct btrfs_root *root, int wait)
mutex_lock(&root->fs_info->ordered_operations_mutex);
spin_lock(&root->fs_info->ordered_extent_lock);
-again:
- list_splice_init(&root->fs_info->ordered_operations, &splice);
-
+ list_splice_init(&cur_trans->ordered_operations, &splice);
while (!list_empty(&splice)) {
-
btrfs_inode = list_entry(splice.next, struct btrfs_inode,
ordered_operations);
-
inode = &btrfs_inode->vfs_inode;
list_del_init(&btrfs_inode->ordered_operations);
@@ -574,24 +642,22 @@ again:
* the inode may be getting freed (in sys_unlink path).
*/
inode = igrab(inode);
-
- if (!wait && inode) {
- list_add_tail(&BTRFS_I(inode)->ordered_operations,
- &root->fs_info->ordered_operations);
- }
-
if (!inode)
continue;
+
+ if (!wait)
+ list_add_tail(&BTRFS_I(inode)->ordered_operations,
+ &cur_trans->ordered_operations);
spin_unlock(&root->fs_info->ordered_extent_lock);
work = btrfs_alloc_delalloc_work(inode, wait, 1);
if (!work) {
+ spin_lock(&root->fs_info->ordered_extent_lock);
if (list_empty(&BTRFS_I(inode)->ordered_operations))
list_add_tail(&btrfs_inode->ordered_operations,
&splice);
- spin_lock(&root->fs_info->ordered_extent_lock);
list_splice_tail(&splice,
- &root->fs_info->ordered_operations);
+ &cur_trans->ordered_operations);
spin_unlock(&root->fs_info->ordered_extent_lock);
ret = -ENOMEM;
goto out;
@@ -603,9 +669,6 @@ again:
cond_resched();
spin_lock(&root->fs_info->ordered_extent_lock);
}
- if (wait && !list_empty(&root->fs_info->ordered_operations))
- goto again;
-
spin_unlock(&root->fs_info->ordered_extent_lock);
out:
list_for_each_entry_safe(work, next, &works, list) {
@@ -923,7 +986,7 @@ out:
* be reclaimed before their checksum is actually put into the btree
*/
int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
- u32 *sum)
+ u32 *sum, int len)
{
struct btrfs_ordered_sum *ordered_sum;
struct btrfs_sector_sum *sector_sums;
@@ -932,22 +995,28 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
unsigned long num_sectors;
unsigned long i;
u32 sectorsize = BTRFS_I(inode)->root->sectorsize;
- int ret = 1;
+ int index = 0;
ordered = btrfs_lookup_ordered_extent(inode, offset);
if (!ordered)
- return 1;
+ return 0;
spin_lock_irq(&tree->lock);
list_for_each_entry_reverse(ordered_sum, &ordered->list, list) {
- if (disk_bytenr >= ordered_sum->bytenr) {
- num_sectors = ordered_sum->len / sectorsize;
- sector_sums = ordered_sum->sums;
- for (i = 0; i < num_sectors; i++) {
+ if (disk_bytenr >= ordered_sum->bytenr &&
+ disk_bytenr < ordered_sum->bytenr + ordered_sum->len) {
+ i = (disk_bytenr - ordered_sum->bytenr) >>
+ inode->i_sb->s_blocksize_bits;
+ sector_sums = ordered_sum->sums + i;
+ num_sectors = ordered_sum->len >>
+ inode->i_sb->s_blocksize_bits;
+ for (; i < num_sectors; i++) {
if (sector_sums[i].bytenr == disk_bytenr) {
- *sum = sector_sums[i].sum;
- ret = 0;
- goto out;
+ sum[index] = sector_sums[i].sum;
+ index++;
+ if (index == len)
+ goto out;
+ disk_bytenr += sectorsize;
}
}
}
@@ -955,7 +1024,7 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
out:
spin_unlock_irq(&tree->lock);
btrfs_put_ordered_extent(ordered);
- return ret;
+ return index;
}
@@ -974,6 +1043,7 @@ out:
void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode)
{
+ struct btrfs_transaction *cur_trans = trans->transaction;
u64 last_mod;
last_mod = max(BTRFS_I(inode)->generation, BTRFS_I(inode)->last_trans);
@@ -988,7 +1058,7 @@ void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
spin_lock(&root->fs_info->ordered_extent_lock);
if (list_empty(&BTRFS_I(inode)->ordered_operations)) {
list_add_tail(&BTRFS_I(inode)->ordered_operations,
- &root->fs_info->ordered_operations);
+ &cur_trans->ordered_operations);
}
spin_unlock(&root->fs_info->ordered_extent_lock);
}
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index f29d4bf5fbe..58b0e3b0eba 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -79,6 +79,8 @@ struct btrfs_ordered_sum {
#define BTRFS_ORDERED_UPDATED_ISIZE 7 /* indicates whether this ordered extent
* has done its due diligence in updating
* the isize. */
+#define BTRFS_ORDERED_LOGGED_CSUM 8 /* We've logged the csums on this ordered
+ ordered extent */
struct btrfs_ordered_extent {
/* logical offset in the file */
@@ -96,6 +98,9 @@ struct btrfs_ordered_extent {
/* number of bytes that still need writing */
u64 bytes_left;
+ /* number of bytes that still need csumming */
+ u64 csum_bytes_left;
+
/*
* the end of the ordered extent which is behind it but
* didn't update disk_i_size. Please see the comment of
@@ -118,6 +123,9 @@ struct btrfs_ordered_extent {
/* list of checksums for insertion when the extent io is done */
struct list_head list;
+ /* If we need to wait on this to be done */
+ struct list_head log_list;
+
/* used to wait for the BTRFS_ORDERED_COMPLETE bit */
wait_queue_head_t wait;
@@ -188,12 +196,17 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
u64 len);
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 btrfs_run_ordered_operations(struct btrfs_root *root, int wait);
+int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
+ u32 *sum, int len);
+int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, int wait);
void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct inode *inode);
void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput);
+void btrfs_get_logged_extents(struct btrfs_root *log, struct inode *inode);
+void btrfs_wait_logged_extents(struct btrfs_root *log, u64 transid);
+void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid);
int __init ordered_data_init(void);
void ordered_data_exit(void);
#endif
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 50d95fd190a..dc0024f17c1 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -176,7 +176,7 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
nr = btrfs_header_nritems(l);
- printk(KERN_INFO "leaf %llu total ptrs %d free space %d\n",
+ btrfs_info(root->fs_info, "leaf %llu total ptrs %d free space %d",
(unsigned long long)btrfs_header_bytenr(l), nr,
btrfs_leaf_free_space(root, l));
for (i = 0 ; i < nr ; i++) {
@@ -294,6 +294,7 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
btrfs_dev_extent_chunk_offset(l, dev_extent),
(unsigned long long)
btrfs_dev_extent_length(l, dev_extent));
+ break;
case BTRFS_DEV_STATS_KEY:
printk(KERN_INFO "\t\tdevice stats\n");
break;
@@ -318,10 +319,9 @@ void btrfs_print_tree(struct btrfs_root *root, struct extent_buffer *c)
btrfs_print_leaf(root, c);
return;
}
- printk(KERN_INFO "node %llu level %d total ptrs %d free spc %u\n",
- (unsigned long long)btrfs_header_bytenr(c),
- level, nr,
- (u32)BTRFS_NODEPTRS_PER_BLOCK(root) - nr);
+ btrfs_info(root->fs_info, "node %llu level %d total ptrs %d free spc %u",
+ (unsigned long long)btrfs_header_bytenr(c),
+ level, nr, (u32)BTRFS_NODEPTRS_PER_BLOCK(root) - nr);
for (i = 0; i < nr; i++) {
btrfs_node_key_to_cpu(c, &key, i);
printk(KERN_INFO "\tkey %d (%llu %u %llu) block %llu\n",
diff --git a/fs/btrfs/print-tree.h b/fs/btrfs/print-tree.h
index da75efe534d..7faddfacc5b 100644
--- a/fs/btrfs/print-tree.h
+++ b/fs/btrfs/print-tree.h
@@ -19,5 +19,5 @@
#ifndef __PRINT_TREE_
#define __PRINT_TREE_
void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l);
-void btrfs_print_tree(struct btrfs_root *root, struct extent_buffer *t);
+void btrfs_print_tree(struct btrfs_root *root, struct extent_buffer *c);
#endif
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index a5c85623432..9d49c586995 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -23,21 +23,21 @@
#include <linux/rbtree.h>
#include <linux/slab.h>
#include <linux/workqueue.h>
+#include <linux/btrfs.h>
#include "ctree.h"
#include "transaction.h"
#include "disk-io.h"
#include "locking.h"
#include "ulist.h"
-#include "ioctl.h"
#include "backref.h"
+#include "extent_io.h"
/* TODO XXX FIXME
* - subvol delete -> delete when ref goes to 0? delete limits also?
* - reorganize keys
* - compressed
* - sync
- * - rescan
* - copy also limits on subvol creation
* - limit
* - caches fuer ulists
@@ -98,7 +98,15 @@ struct btrfs_qgroup_list {
struct btrfs_qgroup *member;
};
-/* must be called with qgroup_lock held */
+struct qgroup_rescan {
+ struct btrfs_work work;
+ struct btrfs_fs_info *fs_info;
+};
+
+static void qgroup_rescan_start(struct btrfs_fs_info *fs_info,
+ struct qgroup_rescan *qscan);
+
+/* must be called with qgroup_ioctl_lock held */
static struct btrfs_qgroup *find_qgroup_rb(struct btrfs_fs_info *fs_info,
u64 qgroupid)
{
@@ -298,7 +306,20 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
}
fs_info->qgroup_flags = btrfs_qgroup_status_flags(l,
ptr);
- /* FIXME read scan element */
+ fs_info->qgroup_rescan_progress.objectid =
+ btrfs_qgroup_status_rescan(l, ptr);
+ if (fs_info->qgroup_flags &
+ BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+ struct qgroup_rescan *qscan =
+ kmalloc(sizeof(*qscan), GFP_NOFS);
+ if (!qscan) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ fs_info->qgroup_rescan_progress.type = 0;
+ fs_info->qgroup_rescan_progress.offset = 0;
+ qgroup_rescan_start(fs_info, qscan);
+ }
goto next1;
}
@@ -420,8 +441,6 @@ void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info)
qgroup = rb_entry(n, struct btrfs_qgroup, node);
rb_erase(n, &fs_info->qgroup_tree);
- WARN_ON(!list_empty(&qgroup->dirty));
-
while (!list_empty(&qgroup->groups)) {
list = list_first_entry(&qgroup->groups,
struct btrfs_qgroup_list,
@@ -620,7 +639,9 @@ static int update_qgroup_limit_item(struct btrfs_trans_handle *trans,
key.offset = qgroupid;
path = btrfs_alloc_path();
- BUG_ON(!path);
+ if (!path)
+ return -ENOMEM;
+
ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
if (ret > 0)
ret = -ENOENT;
@@ -661,7 +682,9 @@ static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
key.offset = qgroup->qgroupid;
path = btrfs_alloc_path();
- BUG_ON(!path);
+ if (!path)
+ return -ENOMEM;
+
ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
if (ret > 0)
ret = -ENOENT;
@@ -702,7 +725,9 @@ static int update_qgroup_status_item(struct btrfs_trans_handle *trans,
key.offset = 0;
path = btrfs_alloc_path();
- BUG_ON(!path);
+ if (!path)
+ return -ENOMEM;
+
ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
if (ret > 0)
ret = -ENOENT;
@@ -715,7 +740,8 @@ static int update_qgroup_status_item(struct btrfs_trans_handle *trans,
ptr = btrfs_item_ptr(l, slot, struct btrfs_qgroup_status_item);
btrfs_set_qgroup_status_flags(l, ptr, fs_info->qgroup_flags);
btrfs_set_qgroup_status_generation(l, ptr, trans->transid);
- /* XXX scan */
+ btrfs_set_qgroup_status_rescan(l, ptr,
+ fs_info->qgroup_rescan_progress.objectid);
btrfs_mark_buffer_dirty(l);
@@ -732,33 +758,38 @@ static int btrfs_clean_quota_tree(struct btrfs_trans_handle *trans,
{
struct btrfs_path *path;
struct btrfs_key key;
+ struct extent_buffer *leaf = NULL;
int ret;
-
- if (!root)
- return -EINVAL;
+ int nr = 0;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
- while (1) {
- key.objectid = 0;
- key.offset = 0;
- key.type = 0;
+ path->leave_spinning = 1;
+
+ key.objectid = 0;
+ key.offset = 0;
+ key.type = 0;
- path->leave_spinning = 1;
+ while (1) {
ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
- if (ret > 0) {
- if (path->slots[0] == 0)
- break;
- path->slots[0]--;
- } else if (ret < 0) {
+ if (ret < 0)
+ goto out;
+ leaf = path->nodes[0];
+ nr = btrfs_header_nritems(leaf);
+ if (!nr)
break;
- }
-
- ret = btrfs_del_item(trans, root, path);
+ /*
+ * delete the leaf one by one
+ * since the whole tree is going
+ * to be deleted.
+ */
+ path->slots[0] = 0;
+ ret = btrfs_del_items(trans, root, path, 0, nr);
if (ret)
goto out;
+
btrfs_release_path(path);
}
ret = 0;
@@ -772,19 +803,21 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info)
{
struct btrfs_root *quota_root;
+ struct btrfs_root *tree_root = fs_info->tree_root;
struct btrfs_path *path = NULL;
struct btrfs_qgroup_status_item *ptr;
struct extent_buffer *leaf;
struct btrfs_key key;
+ struct btrfs_key found_key;
+ struct btrfs_qgroup *qgroup = NULL;
int ret = 0;
+ int slot;
- spin_lock(&fs_info->qgroup_lock);
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
if (fs_info->quota_root) {
fs_info->pending_quota_state = 1;
- spin_unlock(&fs_info->qgroup_lock);
goto out;
}
- spin_unlock(&fs_info->qgroup_lock);
/*
* initially create the quota tree
@@ -819,10 +852,57 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
fs_info->qgroup_flags = BTRFS_QGROUP_STATUS_FLAG_ON |
BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
btrfs_set_qgroup_status_flags(leaf, ptr, fs_info->qgroup_flags);
- btrfs_set_qgroup_status_scan(leaf, ptr, 0);
+ btrfs_set_qgroup_status_rescan(leaf, ptr, 0);
btrfs_mark_buffer_dirty(leaf);
+ key.objectid = 0;
+ key.type = BTRFS_ROOT_REF_KEY;
+ key.offset = 0;
+
+ btrfs_release_path(path);
+ ret = btrfs_search_slot_for_read(tree_root, &key, path, 1, 0);
+ if (ret > 0)
+ goto out_add_root;
+ if (ret < 0)
+ goto out_free_path;
+
+
+ while (1) {
+ slot = path->slots[0];
+ leaf = path->nodes[0];
+ btrfs_item_key_to_cpu(leaf, &found_key, slot);
+
+ if (found_key.type == BTRFS_ROOT_REF_KEY) {
+ ret = add_qgroup_item(trans, quota_root,
+ found_key.offset);
+ if (ret)
+ goto out_free_path;
+
+ qgroup = add_qgroup_rb(fs_info, found_key.offset);
+ if (IS_ERR(qgroup)) {
+ ret = PTR_ERR(qgroup);
+ goto out_free_path;
+ }
+ }
+ ret = btrfs_next_item(tree_root, path);
+ if (ret < 0)
+ goto out_free_path;
+ if (ret)
+ break;
+ }
+
+out_add_root:
+ btrfs_release_path(path);
+ ret = add_qgroup_item(trans, quota_root, BTRFS_FS_TREE_OBJECTID);
+ if (ret)
+ goto out_free_path;
+
+ qgroup = add_qgroup_rb(fs_info, BTRFS_FS_TREE_OBJECTID);
+ if (IS_ERR(qgroup)) {
+ ret = PTR_ERR(qgroup);
+ goto out_free_path;
+ }
spin_lock(&fs_info->qgroup_lock);
fs_info->quota_root = quota_root;
fs_info->pending_quota_state = 1;
@@ -836,6 +916,7 @@ out_free_root:
kfree(quota_root);
}
out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -846,6 +927,9 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
struct btrfs_root *quota_root;
int ret = 0;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
+ if (!fs_info->quota_root)
+ goto out;
spin_lock(&fs_info->qgroup_lock);
fs_info->quota_enabled = 0;
fs_info->pending_quota_state = 0;
@@ -854,8 +938,10 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
btrfs_free_qgroup_config(fs_info);
spin_unlock(&fs_info->qgroup_lock);
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
ret = btrfs_clean_quota_tree(trans, quota_root);
if (ret)
@@ -876,39 +962,62 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
free_extent_buffer(quota_root->commit_root);
kfree(quota_root);
out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
-int btrfs_quota_rescan(struct btrfs_fs_info *fs_info)
+static void qgroup_dirty(struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup *qgroup)
{
- /* FIXME */
- return 0;
+ if (list_empty(&qgroup->dirty))
+ list_add(&qgroup->dirty, &fs_info->dirty_qgroups);
}
int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 src, u64 dst)
{
struct btrfs_root *quota_root;
+ struct btrfs_qgroup *parent;
+ struct btrfs_qgroup *member;
+ struct btrfs_qgroup_list *list;
int ret = 0;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
quota_root = fs_info->quota_root;
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
+ member = find_qgroup_rb(fs_info, src);
+ parent = find_qgroup_rb(fs_info, dst);
+ if (!member || !parent) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ /* check if such qgroup relation exist firstly */
+ list_for_each_entry(list, &member->groups, next_group) {
+ if (list->group == parent) {
+ ret = -EEXIST;
+ goto out;
+ }
+ }
ret = add_qgroup_relation_item(trans, quota_root, src, dst);
if (ret)
- return ret;
+ goto out;
ret = add_qgroup_relation_item(trans, quota_root, dst, src);
if (ret) {
del_qgroup_relation_item(trans, quota_root, src, dst);
- return ret;
+ goto out;
}
spin_lock(&fs_info->qgroup_lock);
ret = add_relation_rb(quota_root->fs_info, src, dst);
spin_unlock(&fs_info->qgroup_lock);
-
+out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -916,13 +1025,34 @@ int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 src, u64 dst)
{
struct btrfs_root *quota_root;
+ struct btrfs_qgroup *parent;
+ struct btrfs_qgroup *member;
+ struct btrfs_qgroup_list *list;
int ret = 0;
int err;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
quota_root = fs_info->quota_root;
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
+ member = find_qgroup_rb(fs_info, src);
+ parent = find_qgroup_rb(fs_info, dst);
+ if (!member || !parent) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ /* check if such qgroup relation exist firstly */
+ list_for_each_entry(list, &member->groups, next_group) {
+ if (list->group == parent)
+ goto exist;
+ }
+ ret = -ENOENT;
+ goto out;
+exist:
ret = del_qgroup_relation_item(trans, quota_root, src, dst);
err = del_qgroup_relation_item(trans, quota_root, dst, src);
if (err && !ret)
@@ -930,9 +1060,9 @@ int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
spin_lock(&fs_info->qgroup_lock);
del_relation_rb(fs_info, src, dst);
-
spin_unlock(&fs_info->qgroup_lock);
-
+out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -943,11 +1073,21 @@ int btrfs_create_qgroup(struct btrfs_trans_handle *trans,
struct btrfs_qgroup *qgroup;
int ret = 0;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
quota_root = fs_info->quota_root;
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
+ qgroup = find_qgroup_rb(fs_info, qgroupid);
+ if (qgroup) {
+ ret = -EEXIST;
+ goto out;
+ }
ret = add_qgroup_item(trans, quota_root, qgroupid);
+ if (ret)
+ goto out;
spin_lock(&fs_info->qgroup_lock);
qgroup = add_qgroup_rb(fs_info, qgroupid);
@@ -955,7 +1095,8 @@ int btrfs_create_qgroup(struct btrfs_trans_handle *trans,
if (IS_ERR(qgroup))
ret = PTR_ERR(qgroup);
-
+out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -966,27 +1107,32 @@ int btrfs_remove_qgroup(struct btrfs_trans_handle *trans,
struct btrfs_qgroup *qgroup;
int ret = 0;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
quota_root = fs_info->quota_root;
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
- /* check if there are no relations to this qgroup */
- spin_lock(&fs_info->qgroup_lock);
qgroup = find_qgroup_rb(fs_info, qgroupid);
- if (qgroup) {
- if (!list_empty(&qgroup->groups) || !list_empty(&qgroup->members)) {
- spin_unlock(&fs_info->qgroup_lock);
- return -EBUSY;
+ if (!qgroup) {
+ ret = -ENOENT;
+ goto out;
+ } else {
+ /* check if there are no relations to this qgroup */
+ if (!list_empty(&qgroup->groups) ||
+ !list_empty(&qgroup->members)) {
+ ret = -EBUSY;
+ goto out;
}
}
- spin_unlock(&fs_info->qgroup_lock);
-
ret = del_qgroup_item(trans, quota_root, qgroupid);
spin_lock(&fs_info->qgroup_lock);
del_qgroup_rb(quota_root->fs_info, qgroupid);
spin_unlock(&fs_info->qgroup_lock);
-
+out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -994,13 +1140,22 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 qgroupid,
struct btrfs_qgroup_limit *limit)
{
- struct btrfs_root *quota_root = fs_info->quota_root;
+ struct btrfs_root *quota_root;
struct btrfs_qgroup *qgroup;
int ret = 0;
- if (!quota_root)
- return -EINVAL;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
+ quota_root = fs_info->quota_root;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
+ qgroup = find_qgroup_rb(fs_info, qgroupid);
+ if (!qgroup) {
+ ret = -ENOENT;
+ goto out;
+ }
ret = update_qgroup_limit_item(trans, quota_root, qgroupid,
limit->flags, limit->max_rfer,
limit->max_excl, limit->rsv_rfer,
@@ -1012,31 +1167,17 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
}
spin_lock(&fs_info->qgroup_lock);
-
- qgroup = find_qgroup_rb(fs_info, qgroupid);
- if (!qgroup) {
- ret = -ENOENT;
- goto unlock;
- }
qgroup->lim_flags = limit->flags;
qgroup->max_rfer = limit->max_rfer;
qgroup->max_excl = limit->max_excl;
qgroup->rsv_rfer = limit->rsv_rfer;
qgroup->rsv_excl = limit->rsv_excl;
-
-unlock:
spin_unlock(&fs_info->qgroup_lock);
-
+out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
-static void qgroup_dirty(struct btrfs_fs_info *fs_info,
- struct btrfs_qgroup *qgroup)
-{
- if (list_empty(&qgroup->dirty))
- list_add(&qgroup->dirty, &fs_info->dirty_qgroups);
-}
-
/*
* btrfs_qgroup_record_ref is called when the ref is added or deleted. it puts
* the modification into a list that's later used by btrfs_end_transaction to
@@ -1060,6 +1201,144 @@ int btrfs_qgroup_record_ref(struct btrfs_trans_handle *trans,
return 0;
}
+static int qgroup_account_ref_step1(struct btrfs_fs_info *fs_info,
+ struct ulist *roots, struct ulist *tmp,
+ u64 seq)
+{
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+ struct ulist_node *tmp_unode;
+ struct ulist_iterator tmp_uiter;
+ struct btrfs_qgroup *qg;
+ int ret;
+
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(roots, &uiter))) {
+ qg = find_qgroup_rb(fs_info, unode->val);
+ if (!qg)
+ continue;
+
+ ulist_reinit(tmp);
+ /* XXX id not needed */
+ ret = ulist_add(tmp, qg->qgroupid,
+ (u64)(uintptr_t)qg, GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+ ULIST_ITER_INIT(&tmp_uiter);
+ while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
+ struct btrfs_qgroup_list *glist;
+
+ qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
+ if (qg->refcnt < seq)
+ qg->refcnt = seq + 1;
+ else
+ ++qg->refcnt;
+
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ ret = ulist_add(tmp, glist->group->qgroupid,
+ (u64)(uintptr_t)glist->group,
+ GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+ }
+ }
+ }
+
+ return 0;
+}
+
+static int qgroup_account_ref_step2(struct btrfs_fs_info *fs_info,
+ struct ulist *roots, struct ulist *tmp,
+ u64 seq, int sgn, u64 num_bytes,
+ struct btrfs_qgroup *qgroup)
+{
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+ struct btrfs_qgroup *qg;
+ struct btrfs_qgroup_list *glist;
+ int ret;
+
+ ulist_reinit(tmp);
+ ret = ulist_add(tmp, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(tmp, &uiter))) {
+ qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
+ if (qg->refcnt < seq) {
+ /* not visited by step 1 */
+ qg->rfer += sgn * num_bytes;
+ qg->rfer_cmpr += sgn * num_bytes;
+ if (roots->nnodes == 0) {
+ qg->excl += sgn * num_bytes;
+ qg->excl_cmpr += sgn * num_bytes;
+ }
+ qgroup_dirty(fs_info, qg);
+ }
+ WARN_ON(qg->tag >= seq);
+ qg->tag = seq;
+
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ ret = ulist_add(tmp, glist->group->qgroupid,
+ (uintptr_t)glist->group, GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+ }
+ }
+
+ return 0;
+}
+
+static int qgroup_account_ref_step3(struct btrfs_fs_info *fs_info,
+ struct ulist *roots, struct ulist *tmp,
+ u64 seq, int sgn, u64 num_bytes)
+{
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+ struct btrfs_qgroup *qg;
+ struct ulist_node *tmp_unode;
+ struct ulist_iterator tmp_uiter;
+ int ret;
+
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(roots, &uiter))) {
+ qg = find_qgroup_rb(fs_info, unode->val);
+ if (!qg)
+ continue;
+
+ ulist_reinit(tmp);
+ ret = ulist_add(tmp, qg->qgroupid, (uintptr_t)qg, GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+
+ ULIST_ITER_INIT(&tmp_uiter);
+ while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
+ struct btrfs_qgroup_list *glist;
+
+ qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
+ if (qg->tag == seq)
+ continue;
+
+ if (qg->refcnt - seq == roots->nnodes) {
+ qg->excl -= sgn * num_bytes;
+ qg->excl_cmpr -= sgn * num_bytes;
+ qgroup_dirty(fs_info, qg);
+ }
+
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ ret = ulist_add(tmp, glist->group->qgroupid,
+ (uintptr_t)glist->group,
+ GFP_ATOMIC);
+ if (ret < 0)
+ return ret;
+ }
+ }
+ }
+
+ return 0;
+}
+
/*
* btrfs_qgroup_account_ref is called for every ref that is added to or deleted
* from the fs. First, all roots referencing the extent are searched, and
@@ -1075,10 +1354,8 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans,
struct btrfs_root *quota_root;
u64 ref_root;
struct btrfs_qgroup *qgroup;
- struct ulist_node *unode;
struct ulist *roots = NULL;
struct ulist *tmp = NULL;
- struct ulist_iterator uiter;
u64 seq;
int ret = 0;
int sgn;
@@ -1117,9 +1394,11 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans,
case BTRFS_ADD_DELAYED_REF:
case BTRFS_ADD_DELAYED_EXTENT:
sgn = 1;
+ seq = btrfs_tree_mod_seq_prev(node->seq);
break;
case BTRFS_DROP_DELAYED_REF:
sgn = -1;
+ seq = node->seq;
break;
case BTRFS_UPDATE_DELAYED_HEAD:
return 0;
@@ -1127,20 +1406,37 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans,
BUG();
}
+ mutex_lock(&fs_info->qgroup_rescan_lock);
+ if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+ if (fs_info->qgroup_rescan_progress.objectid <= node->bytenr) {
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+ return 0;
+ }
+ }
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+
/*
* the delayed ref sequence number we pass depends on the direction of
- * the operation. for add operations, we pass (node->seq - 1) to skip
+ * the operation. for add operations, we pass
+ * tree_mod_log_prev_seq(node->seq) to skip
* the delayed ref's current sequence number, because we need the state
* of the tree before the add operation. for delete operations, we pass
* (node->seq) to include the delayed ref's current sequence number,
* because we need the state of the tree after the delete operation.
*/
- ret = btrfs_find_all_roots(trans, fs_info, node->bytenr,
- sgn > 0 ? node->seq - 1 : node->seq, &roots);
+ ret = btrfs_find_all_roots(trans, fs_info, node->bytenr, seq, &roots);
if (ret < 0)
- goto out;
+ return ret;
+ mutex_lock(&fs_info->qgroup_rescan_lock);
spin_lock(&fs_info->qgroup_lock);
+ if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+ if (fs_info->qgroup_rescan_progress.objectid <= node->bytenr) {
+ ret = 0;
+ goto unlock;
+ }
+ }
+
quota_root = fs_info->quota_root;
if (!quota_root)
goto unlock;
@@ -1160,107 +1456,29 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans,
seq = fs_info->qgroup_seq;
fs_info->qgroup_seq += roots->nnodes + 1; /* max refcnt */
- ULIST_ITER_INIT(&uiter);
- while ((unode = ulist_next(roots, &uiter))) {
- struct ulist_node *tmp_unode;
- struct ulist_iterator tmp_uiter;
- struct btrfs_qgroup *qg;
-
- qg = find_qgroup_rb(fs_info, unode->val);
- if (!qg)
- continue;
-
- ulist_reinit(tmp);
- /* XXX id not needed */
- ulist_add(tmp, qg->qgroupid, (u64)(uintptr_t)qg, GFP_ATOMIC);
- ULIST_ITER_INIT(&tmp_uiter);
- while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
- struct btrfs_qgroup_list *glist;
-
- qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
- if (qg->refcnt < seq)
- qg->refcnt = seq + 1;
- else
- ++qg->refcnt;
-
- list_for_each_entry(glist, &qg->groups, next_group) {
- ulist_add(tmp, glist->group->qgroupid,
- (u64)(uintptr_t)glist->group,
- GFP_ATOMIC);
- }
- }
- }
+ ret = qgroup_account_ref_step1(fs_info, roots, tmp, seq);
+ if (ret)
+ goto unlock;
/*
* step 2: walk from the new root
*/
- ulist_reinit(tmp);
- ulist_add(tmp, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
- ULIST_ITER_INIT(&uiter);
- while ((unode = ulist_next(tmp, &uiter))) {
- struct btrfs_qgroup *qg;
- struct btrfs_qgroup_list *glist;
-
- qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
- if (qg->refcnt < seq) {
- /* not visited by step 1 */
- qg->rfer += sgn * node->num_bytes;
- qg->rfer_cmpr += sgn * node->num_bytes;
- if (roots->nnodes == 0) {
- qg->excl += sgn * node->num_bytes;
- qg->excl_cmpr += sgn * node->num_bytes;
- }
- qgroup_dirty(fs_info, qg);
- }
- WARN_ON(qg->tag >= seq);
- qg->tag = seq;
-
- list_for_each_entry(glist, &qg->groups, next_group) {
- ulist_add(tmp, glist->group->qgroupid,
- (uintptr_t)glist->group, GFP_ATOMIC);
- }
- }
+ ret = qgroup_account_ref_step2(fs_info, roots, tmp, seq, sgn,
+ node->num_bytes, qgroup);
+ if (ret)
+ goto unlock;
/*
* step 3: walk again from old refs
*/
- ULIST_ITER_INIT(&uiter);
- while ((unode = ulist_next(roots, &uiter))) {
- struct btrfs_qgroup *qg;
- struct ulist_node *tmp_unode;
- struct ulist_iterator tmp_uiter;
-
- qg = find_qgroup_rb(fs_info, unode->val);
- if (!qg)
- continue;
-
- ulist_reinit(tmp);
- ulist_add(tmp, qg->qgroupid, (uintptr_t)qg, GFP_ATOMIC);
- ULIST_ITER_INIT(&tmp_uiter);
- while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
- struct btrfs_qgroup_list *glist;
-
- qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
- if (qg->tag == seq)
- continue;
-
- if (qg->refcnt - seq == roots->nnodes) {
- qg->excl -= sgn * node->num_bytes;
- qg->excl_cmpr -= sgn * node->num_bytes;
- qgroup_dirty(fs_info, qg);
- }
+ ret = qgroup_account_ref_step3(fs_info, roots, tmp, seq, sgn,
+ node->num_bytes);
+ if (ret)
+ goto unlock;
- list_for_each_entry(glist, &qg->groups, next_group) {
- ulist_add(tmp, glist->group->qgroupid,
- (uintptr_t)glist->group,
- GFP_ATOMIC);
- }
- }
- }
- ret = 0;
unlock:
spin_unlock(&fs_info->qgroup_lock);
-out:
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
ulist_free(roots);
ulist_free(tmp);
@@ -1275,10 +1493,14 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
{
struct btrfs_root *quota_root = fs_info->quota_root;
int ret = 0;
+ int start_rescan_worker = 0;
if (!quota_root)
goto out;
+ if (!fs_info->quota_enabled && fs_info->pending_quota_state)
+ start_rescan_worker = 1;
+
fs_info->quota_enabled = fs_info->pending_quota_state;
spin_lock(&fs_info->qgroup_lock);
@@ -1304,6 +1526,13 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
if (ret)
fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+ if (!ret && start_rescan_worker) {
+ ret = btrfs_qgroup_rescan(fs_info);
+ if (ret)
+ pr_err("btrfs: start rescan quota failed: %d\n", ret);
+ ret = 0;
+ }
+
out:
return ret;
@@ -1324,12 +1553,30 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
struct btrfs_qgroup *srcgroup;
struct btrfs_qgroup *dstgroup;
u32 level_size = 0;
+ u64 nums;
+ mutex_lock(&fs_info->qgroup_ioctl_lock);
if (!fs_info->quota_enabled)
- return 0;
+ goto out;
- if (!quota_root)
- return -EINVAL;
+ if (!quota_root) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (inherit) {
+ i_qgroups = (u64 *)(inherit + 1);
+ nums = inherit->num_qgroups + 2 * inherit->num_ref_copies +
+ 2 * inherit->num_excl_copies;
+ for (i = 0; i < nums; ++i) {
+ srcgroup = find_qgroup_rb(fs_info, *i_qgroups);
+ if (!srcgroup) {
+ ret = -EINVAL;
+ goto out;
+ }
+ ++i_qgroups;
+ }
+ }
/*
* create a tracking group for the subvol itself
@@ -1456,6 +1703,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
unlock:
spin_unlock(&fs_info->qgroup_lock);
out:
+ mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
@@ -1500,7 +1748,10 @@ int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
ret = -ENOMEM;
goto out;
}
- ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
+ ret = ulist_add(ulist, qgroup->qgroupid,
+ (uintptr_t)qgroup, GFP_ATOMIC);
+ if (ret < 0)
+ goto out;
ULIST_ITER_INIT(&uiter);
while ((unode = ulist_next(ulist, &uiter))) {
struct btrfs_qgroup *qg;
@@ -1509,23 +1760,27 @@ int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
- qg->reserved + qg->rfer + num_bytes >
- qg->max_rfer)
+ qg->reserved + (s64)qg->rfer + num_bytes >
+ qg->max_rfer) {
ret = -EDQUOT;
+ goto out;
+ }
if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
- qg->reserved + qg->excl + num_bytes >
- qg->max_excl)
+ qg->reserved + (s64)qg->excl + num_bytes >
+ qg->max_excl) {
ret = -EDQUOT;
+ goto out;
+ }
list_for_each_entry(glist, &qg->groups, next_group) {
- ulist_add(ulist, glist->group->qgroupid,
- (uintptr_t)glist->group, GFP_ATOMIC);
+ ret = ulist_add(ulist, glist->group->qgroupid,
+ (uintptr_t)glist->group, GFP_ATOMIC);
+ if (ret < 0)
+ goto out;
}
}
- if (ret)
- goto out;
-
+ ret = 0;
/*
* no limits exceeded, now record the reservation into all qgroups
*/
@@ -1554,6 +1809,7 @@ void btrfs_qgroup_free(struct btrfs_root *root, u64 num_bytes)
struct ulist_node *unode;
struct ulist_iterator uiter;
u64 ref_root = root->root_key.objectid;
+ int ret = 0;
if (!is_fstree(ref_root))
return;
@@ -1576,7 +1832,10 @@ void btrfs_qgroup_free(struct btrfs_root *root, u64 num_bytes)
btrfs_std_error(fs_info, -ENOMEM);
goto out;
}
- ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
+ ret = ulist_add(ulist, qgroup->qgroupid,
+ (uintptr_t)qgroup, GFP_ATOMIC);
+ if (ret < 0)
+ goto out;
ULIST_ITER_INIT(&uiter);
while ((unode = ulist_next(ulist, &uiter))) {
struct btrfs_qgroup *qg;
@@ -1587,8 +1846,10 @@ void btrfs_qgroup_free(struct btrfs_root *root, u64 num_bytes)
qg->reserved -= num_bytes;
list_for_each_entry(glist, &qg->groups, next_group) {
- ulist_add(ulist, glist->group->qgroupid,
- (uintptr_t)glist->group, GFP_ATOMIC);
+ ret = ulist_add(ulist, glist->group->qgroupid,
+ (uintptr_t)glist->group, GFP_ATOMIC);
+ if (ret < 0)
+ goto out;
}
}
@@ -1601,8 +1862,265 @@ void assert_qgroups_uptodate(struct btrfs_trans_handle *trans)
{
if (list_empty(&trans->qgroup_ref_list) && !trans->delayed_ref_elem.seq)
return;
- printk(KERN_ERR "btrfs: qgroups not uptodate in trans handle %p: list is%s empty, seq is %llu\n",
+ pr_err("btrfs: qgroups not uptodate in trans handle %p: list is%s empty, seq is %#x.%x\n",
trans, list_empty(&trans->qgroup_ref_list) ? "" : " not",
- trans->delayed_ref_elem.seq);
+ (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, 2 when done and FLAG_INCONSISTENT was cleared.
+ */
+static int
+qgroup_rescan_leaf(struct qgroup_rescan *qscan, struct btrfs_path *path,
+ struct btrfs_trans_handle *trans, struct ulist *tmp,
+ struct extent_buffer *scratch_leaf)
+{
+ struct btrfs_key found;
+ struct btrfs_fs_info *fs_info = qscan->fs_info;
+ struct ulist *roots = NULL;
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+ struct seq_list tree_mod_seq_elem = {};
+ u64 seq;
+ int slot;
+ int ret;
+
+ path->leave_spinning = 1;
+ mutex_lock(&fs_info->qgroup_rescan_lock);
+ ret = btrfs_search_slot_for_read(fs_info->extent_root,
+ &fs_info->qgroup_rescan_progress,
+ path, 1, 0);
+
+ pr_debug("current progress key (%llu %u %llu), search_slot ret %d\n",
+ (unsigned long long)fs_info->qgroup_rescan_progress.objectid,
+ fs_info->qgroup_rescan_progress.type,
+ (unsigned long long)fs_info->qgroup_rescan_progress.offset,
+ ret);
+
+ if (ret) {
+ /*
+ * The rescan is about to end, we will not be scanning any
+ * further blocks. We cannot unset the RESCAN flag here, because
+ * we want to commit the transaction if everything went well.
+ * To make the live accounting work in this phase, we set our
+ * scan progress pointer such that every real extent objectid
+ * will be smaller.
+ */
+ fs_info->qgroup_rescan_progress.objectid = (u64)-1;
+ btrfs_release_path(path);
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+ return ret;
+ }
+
+ btrfs_item_key_to_cpu(path->nodes[0], &found,
+ btrfs_header_nritems(path->nodes[0]) - 1);
+ fs_info->qgroup_rescan_progress.objectid = found.objectid + 1;
+
+ btrfs_get_tree_mod_seq(fs_info, &tree_mod_seq_elem);
+ memcpy(scratch_leaf, path->nodes[0], sizeof(*scratch_leaf));
+ slot = path->slots[0];
+ btrfs_release_path(path);
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+
+ for (; slot < btrfs_header_nritems(scratch_leaf); ++slot) {
+ btrfs_item_key_to_cpu(scratch_leaf, &found, slot);
+ if (found.type != BTRFS_EXTENT_ITEM_KEY)
+ continue;
+ ret = btrfs_find_all_roots(trans, fs_info, found.objectid,
+ tree_mod_seq_elem.seq, &roots);
+ if (ret < 0)
+ goto out;
+ spin_lock(&fs_info->qgroup_lock);
+ seq = fs_info->qgroup_seq;
+ fs_info->qgroup_seq += roots->nnodes + 1; /* max refcnt */
+
+ ret = qgroup_account_ref_step1(fs_info, roots, tmp, seq);
+ if (ret) {
+ spin_unlock(&fs_info->qgroup_lock);
+ ulist_free(roots);
+ goto out;
+ }
+
+ /*
+ * step2 of btrfs_qgroup_account_ref works from a single root,
+ * we're doing all at once here.
+ */
+ ulist_reinit(tmp);
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(roots, &uiter))) {
+ struct btrfs_qgroup *qg;
+
+ qg = find_qgroup_rb(fs_info, unode->val);
+ if (!qg)
+ continue;
+
+ ret = ulist_add(tmp, qg->qgroupid, (uintptr_t)qg,
+ GFP_ATOMIC);
+ if (ret < 0) {
+ spin_unlock(&fs_info->qgroup_lock);
+ ulist_free(roots);
+ goto out;
+ }
+ }
+
+ /* this loop is similar to step 2 of btrfs_qgroup_account_ref */
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(tmp, &uiter))) {
+ struct btrfs_qgroup *qg;
+ struct btrfs_qgroup_list *glist;
+
+ qg = (struct btrfs_qgroup *)(uintptr_t) unode->aux;
+ qg->rfer += found.offset;
+ qg->rfer_cmpr += found.offset;
+ WARN_ON(qg->tag >= seq);
+ if (qg->refcnt - seq == roots->nnodes) {
+ qg->excl += found.offset;
+ qg->excl_cmpr += found.offset;
+ }
+ qgroup_dirty(fs_info, qg);
+
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ ret = ulist_add(tmp, glist->group->qgroupid,
+ (uintptr_t)glist->group,
+ GFP_ATOMIC);
+ if (ret < 0) {
+ spin_unlock(&fs_info->qgroup_lock);
+ ulist_free(roots);
+ goto out;
+ }
+ }
+ }
+
+ spin_unlock(&fs_info->qgroup_lock);
+ ulist_free(roots);
+ ret = 0;
+ }
+
+out:
+ btrfs_put_tree_mod_seq(fs_info, &tree_mod_seq_elem);
+
+ return ret;
+}
+
+static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
+{
+ struct qgroup_rescan *qscan = container_of(work, struct qgroup_rescan,
+ work);
+ struct btrfs_path *path;
+ struct btrfs_trans_handle *trans = NULL;
+ struct btrfs_fs_info *fs_info = qscan->fs_info;
+ struct ulist *tmp = NULL;
+ struct extent_buffer *scratch_leaf = NULL;
+ int err = -ENOMEM;
+
+ path = btrfs_alloc_path();
+ if (!path)
+ goto out;
+ tmp = ulist_alloc(GFP_NOFS);
+ if (!tmp)
+ goto out;
+ scratch_leaf = kmalloc(sizeof(*scratch_leaf), GFP_NOFS);
+ if (!scratch_leaf)
+ goto out;
+
+ err = 0;
+ while (!err) {
+ trans = btrfs_start_transaction(fs_info->fs_root, 0);
+ if (IS_ERR(trans)) {
+ err = PTR_ERR(trans);
+ break;
+ }
+ if (!fs_info->quota_enabled) {
+ err = -EINTR;
+ } else {
+ err = qgroup_rescan_leaf(qscan, path, trans,
+ tmp, scratch_leaf);
+ }
+ if (err > 0)
+ btrfs_commit_transaction(trans, fs_info->fs_root);
+ else
+ btrfs_end_transaction(trans, fs_info->fs_root);
+ }
+
+out:
+ kfree(scratch_leaf);
+ ulist_free(tmp);
+ btrfs_free_path(path);
+ kfree(qscan);
+
+ mutex_lock(&fs_info->qgroup_rescan_lock);
+ fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_RESCAN;
+
+ if (err == 2 &&
+ fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT) {
+ fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+ } else if (err < 0) {
+ fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+ }
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+
+ if (err >= 0) {
+ pr_info("btrfs: qgroup scan completed%s\n",
+ err == 2 ? " (inconsistency flag cleared)" : "");
+ } else {
+ pr_err("btrfs: qgroup scan failed with %d\n", err);
+ }
+}
+
+static void
+qgroup_rescan_start(struct btrfs_fs_info *fs_info, struct qgroup_rescan *qscan)
+{
+ memset(&qscan->work, 0, sizeof(qscan->work));
+ qscan->work.func = btrfs_qgroup_rescan_worker;
+ qscan->fs_info = fs_info;
+
+ pr_info("btrfs: qgroup scan started\n");
+ btrfs_queue_worker(&fs_info->qgroup_rescan_workers, &qscan->work);
+}
+
+int
+btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info)
+{
+ int ret = 0;
+ struct rb_node *n;
+ struct btrfs_qgroup *qgroup;
+ struct qgroup_rescan *qscan = kmalloc(sizeof(*qscan), GFP_NOFS);
+
+ if (!qscan)
+ return -ENOMEM;
+
+ mutex_lock(&fs_info->qgroup_rescan_lock);
+ spin_lock(&fs_info->qgroup_lock);
+ if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN)
+ ret = -EINPROGRESS;
+ else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))
+ ret = -EINVAL;
+ if (ret) {
+ spin_unlock(&fs_info->qgroup_lock);
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+ kfree(qscan);
+ return ret;
+ }
+
+ fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_RESCAN;
+ memset(&fs_info->qgroup_rescan_progress, 0,
+ sizeof(fs_info->qgroup_rescan_progress));
+
+ /* clear all current qgroup tracking information */
+ for (n = rb_first(&fs_info->qgroup_tree); n; n = rb_next(n)) {
+ qgroup = rb_entry(n, struct btrfs_qgroup, node);
+ qgroup->rfer = 0;
+ qgroup->rfer_cmpr = 0;
+ qgroup->excl = 0;
+ qgroup->excl_cmpr = 0;
+ }
+ spin_unlock(&fs_info->qgroup_lock);
+ mutex_unlock(&fs_info->qgroup_rescan_lock);
+
+ qgroup_rescan_start(fs_info, qscan);
+
+ return 0;
+}
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
new file mode 100644
index 00000000000..0740621daf6
--- /dev/null
+++ b/fs/btrfs/raid56.c
@@ -0,0 +1,2100 @@
+/*
+ * Copyright (C) 2012 Fusion-io All rights reserved.
+ * Copyright (C) 2012 Intel Corp. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/bio.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+#include <linux/blkdev.h>
+#include <linux/random.h>
+#include <linux/iocontext.h>
+#include <linux/capability.h>
+#include <linux/ratelimit.h>
+#include <linux/kthread.h>
+#include <linux/raid/pq.h>
+#include <linux/hash.h>
+#include <linux/list_sort.h>
+#include <linux/raid/xor.h>
+#include <linux/vmalloc.h>
+#include <asm/div64.h>
+#include "compat.h"
+#include "ctree.h"
+#include "extent_map.h"
+#include "disk-io.h"
+#include "transaction.h"
+#include "print-tree.h"
+#include "volumes.h"
+#include "raid56.h"
+#include "async-thread.h"
+#include "check-integrity.h"
+#include "rcu-string.h"
+
+/* set when additional merges to this rbio are not allowed */
+#define RBIO_RMW_LOCKED_BIT 1
+
+/*
+ * set when this rbio is sitting in the hash, but it is just a cache
+ * of past RMW
+ */
+#define RBIO_CACHE_BIT 2
+
+/*
+ * set when it is safe to trust the stripe_pages for caching
+ */
+#define RBIO_CACHE_READY_BIT 3
+
+
+#define RBIO_CACHE_SIZE 1024
+
+struct btrfs_raid_bio {
+ struct btrfs_fs_info *fs_info;
+ struct btrfs_bio *bbio;
+
+ /*
+ * logical block numbers for the start of each stripe
+ * The last one or two are p/q. These are sorted,
+ * so raid_map[0] is the start of our full stripe
+ */
+ u64 *raid_map;
+
+ /* while we're doing rmw on a stripe
+ * we put it into a hash table so we can
+ * lock the stripe and merge more rbios
+ * into it.
+ */
+ struct list_head hash_list;
+
+ /*
+ * LRU list for the stripe cache
+ */
+ struct list_head stripe_cache;
+
+ /*
+ * for scheduling work in the helper threads
+ */
+ struct btrfs_work work;
+
+ /*
+ * bio list and bio_list_lock are used
+ * to add more bios into the stripe
+ * in hopes of avoiding the full rmw
+ */
+ struct bio_list bio_list;
+ spinlock_t bio_list_lock;
+
+ /* also protected by the bio_list_lock, the
+ * plug list is used by the plugging code
+ * to collect partial bios while plugged. The
+ * stripe locking code also uses it to hand off
+ * the stripe lock to the next pending IO
+ */
+ struct list_head plug_list;
+
+ /*
+ * flags that tell us if it is safe to
+ * merge with this bio
+ */
+ unsigned long flags;
+
+ /* size of each individual stripe on disk */
+ int stripe_len;
+
+ /* number of data stripes (no p/q) */
+ int nr_data;
+
+ /*
+ * set if we're doing a parity rebuild
+ * for a read from higher up, which is handled
+ * differently from a parity rebuild as part of
+ * rmw
+ */
+ int read_rebuild;
+
+ /* first bad stripe */
+ int faila;
+
+ /* second bad stripe (for raid6 use) */
+ int failb;
+
+ /*
+ * number of pages needed to represent the full
+ * stripe
+ */
+ int nr_pages;
+
+ /*
+ * size of all the bios in the bio_list. This
+ * helps us decide if the rbio maps to a full
+ * stripe or not
+ */
+ int bio_list_bytes;
+
+ atomic_t refs;
+
+ /*
+ * these are two arrays of pointers. We allocate the
+ * rbio big enough to hold them both and setup their
+ * locations when the rbio is allocated
+ */
+
+ /* pointers to pages that we allocated for
+ * reading/writing stripes directly from the disk (including P/Q)
+ */
+ struct page **stripe_pages;
+
+ /*
+ * pointers to the pages in the bio_list. Stored
+ * here for faster lookup
+ */
+ struct page **bio_pages;
+};
+
+static int __raid56_parity_recover(struct btrfs_raid_bio *rbio);
+static noinline void finish_rmw(struct btrfs_raid_bio *rbio);
+static void rmw_work(struct btrfs_work *work);
+static void read_rebuild_work(struct btrfs_work *work);
+static void async_rmw_stripe(struct btrfs_raid_bio *rbio);
+static void async_read_rebuild(struct btrfs_raid_bio *rbio);
+static int fail_bio_stripe(struct btrfs_raid_bio *rbio, struct bio *bio);
+static int fail_rbio_index(struct btrfs_raid_bio *rbio, int failed);
+static void __free_raid_bio(struct btrfs_raid_bio *rbio);
+static void index_rbio_pages(struct btrfs_raid_bio *rbio);
+static int alloc_rbio_pages(struct btrfs_raid_bio *rbio);
+
+/*
+ * the stripe hash table is used for locking, and to collect
+ * bios in hopes of making a full stripe
+ */
+int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
+{
+ struct btrfs_stripe_hash_table *table;
+ struct btrfs_stripe_hash_table *x;
+ struct btrfs_stripe_hash *cur;
+ struct btrfs_stripe_hash *h;
+ int num_entries = 1 << BTRFS_STRIPE_HASH_TABLE_BITS;
+ int i;
+ int table_size;
+
+ if (info->stripe_hash_table)
+ return 0;
+
+ /*
+ * The table is large, starting with order 4 and can go as high as
+ * order 7 in case lock debugging is turned on.
+ *
+ * Try harder to allocate and fallback to vmalloc to lower the chance
+ * 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;
+ }
+
+ spin_lock_init(&table->cache_lock);
+ INIT_LIST_HEAD(&table->stripe_cache);
+
+ h = table->table;
+
+ for (i = 0; i < num_entries; i++) {
+ cur = h + i;
+ INIT_LIST_HEAD(&cur->hash_list);
+ spin_lock_init(&cur->lock);
+ init_waitqueue_head(&cur->wait);
+ }
+
+ x = cmpxchg(&info->stripe_hash_table, NULL, table);
+ if (x) {
+ if (is_vmalloc_addr(x))
+ vfree(x);
+ else
+ kfree(x);
+ }
+ return 0;
+}
+
+/*
+ * caching an rbio means to copy anything from the
+ * bio_pages array into the stripe_pages array. We
+ * use the page uptodate bit in the stripe cache array
+ * to indicate if it has valid data
+ *
+ * once the caching is done, we set the cache ready
+ * bit.
+ */
+static void cache_rbio_pages(struct btrfs_raid_bio *rbio)
+{
+ int i;
+ char *s;
+ char *d;
+ int ret;
+
+ ret = alloc_rbio_pages(rbio);
+ if (ret)
+ return;
+
+ for (i = 0; i < rbio->nr_pages; i++) {
+ if (!rbio->bio_pages[i])
+ continue;
+
+ s = kmap(rbio->bio_pages[i]);
+ d = kmap(rbio->stripe_pages[i]);
+
+ memcpy(d, s, PAGE_CACHE_SIZE);
+
+ kunmap(rbio->bio_pages[i]);
+ kunmap(rbio->stripe_pages[i]);
+ SetPageUptodate(rbio->stripe_pages[i]);
+ }
+ set_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
+}
+
+/*
+ * we hash on the first logical address of the stripe
+ */
+static int rbio_bucket(struct btrfs_raid_bio *rbio)
+{
+ u64 num = rbio->raid_map[0];
+
+ /*
+ * we shift down quite a bit. We're using byte
+ * addressing, and most of the lower bits are zeros.
+ * This tends to upset hash_64, and it consistently
+ * returns just one or two different values.
+ *
+ * shifting off the lower bits fixes things.
+ */
+ return hash_64(num >> 16, BTRFS_STRIPE_HASH_TABLE_BITS);
+}
+
+/*
+ * stealing an rbio means taking all the uptodate pages from the stripe
+ * array in the source rbio and putting them into the destination rbio
+ */
+static void steal_rbio(struct btrfs_raid_bio *src, struct btrfs_raid_bio *dest)
+{
+ int i;
+ struct page *s;
+ struct page *d;
+
+ if (!test_bit(RBIO_CACHE_READY_BIT, &src->flags))
+ return;
+
+ for (i = 0; i < dest->nr_pages; i++) {
+ s = src->stripe_pages[i];
+ if (!s || !PageUptodate(s)) {
+ continue;
+ }
+
+ d = dest->stripe_pages[i];
+ if (d)
+ __free_page(d);
+
+ dest->stripe_pages[i] = s;
+ src->stripe_pages[i] = NULL;
+ }
+}
+
+/*
+ * merging means we take the bio_list from the victim and
+ * splice it into the destination. The victim should
+ * be discarded afterwards.
+ *
+ * must be called with dest->rbio_list_lock held
+ */
+static void merge_rbio(struct btrfs_raid_bio *dest,
+ struct btrfs_raid_bio *victim)
+{
+ bio_list_merge(&dest->bio_list, &victim->bio_list);
+ dest->bio_list_bytes += victim->bio_list_bytes;
+ bio_list_init(&victim->bio_list);
+}
+
+/*
+ * used to prune items that are in the cache. The caller
+ * must hold the hash table lock.
+ */
+static void __remove_rbio_from_cache(struct btrfs_raid_bio *rbio)
+{
+ int bucket = rbio_bucket(rbio);
+ struct btrfs_stripe_hash_table *table;
+ struct btrfs_stripe_hash *h;
+ int freeit = 0;
+
+ /*
+ * check the bit again under the hash table lock.
+ */
+ if (!test_bit(RBIO_CACHE_BIT, &rbio->flags))
+ return;
+
+ table = rbio->fs_info->stripe_hash_table;
+ h = table->table + bucket;
+
+ /* hold the lock for the bucket because we may be
+ * removing it from the hash table
+ */
+ spin_lock(&h->lock);
+
+ /*
+ * hold the lock for the bio list because we need
+ * to make sure the bio list is empty
+ */
+ spin_lock(&rbio->bio_list_lock);
+
+ if (test_and_clear_bit(RBIO_CACHE_BIT, &rbio->flags)) {
+ list_del_init(&rbio->stripe_cache);
+ table->cache_size -= 1;
+ freeit = 1;
+
+ /* if the bio list isn't empty, this rbio is
+ * still involved in an IO. We take it out
+ * of the cache list, and drop the ref that
+ * was held for the list.
+ *
+ * If the bio_list was empty, we also remove
+ * the rbio from the hash_table, and drop
+ * the corresponding ref
+ */
+ if (bio_list_empty(&rbio->bio_list)) {
+ if (!list_empty(&rbio->hash_list)) {
+ list_del_init(&rbio->hash_list);
+ atomic_dec(&rbio->refs);
+ BUG_ON(!list_empty(&rbio->plug_list));
+ }
+ }
+ }
+
+ spin_unlock(&rbio->bio_list_lock);
+ spin_unlock(&h->lock);
+
+ if (freeit)
+ __free_raid_bio(rbio);
+}
+
+/*
+ * prune a given rbio from the cache
+ */
+static void remove_rbio_from_cache(struct btrfs_raid_bio *rbio)
+{
+ struct btrfs_stripe_hash_table *table;
+ unsigned long flags;
+
+ if (!test_bit(RBIO_CACHE_BIT, &rbio->flags))
+ return;
+
+ table = rbio->fs_info->stripe_hash_table;
+
+ spin_lock_irqsave(&table->cache_lock, flags);
+ __remove_rbio_from_cache(rbio);
+ spin_unlock_irqrestore(&table->cache_lock, flags);
+}
+
+/*
+ * remove everything in the cache
+ */
+static void btrfs_clear_rbio_cache(struct btrfs_fs_info *info)
+{
+ struct btrfs_stripe_hash_table *table;
+ unsigned long flags;
+ struct btrfs_raid_bio *rbio;
+
+ table = info->stripe_hash_table;
+
+ spin_lock_irqsave(&table->cache_lock, flags);
+ while (!list_empty(&table->stripe_cache)) {
+ rbio = list_entry(table->stripe_cache.next,
+ struct btrfs_raid_bio,
+ stripe_cache);
+ __remove_rbio_from_cache(rbio);
+ }
+ spin_unlock_irqrestore(&table->cache_lock, flags);
+}
+
+/*
+ * remove all cached entries and free the hash table
+ * used by unmount
+ */
+void btrfs_free_stripe_hash_table(struct btrfs_fs_info *info)
+{
+ if (!info->stripe_hash_table)
+ return;
+ btrfs_clear_rbio_cache(info);
+ if (is_vmalloc_addr(info->stripe_hash_table))
+ vfree(info->stripe_hash_table);
+ else
+ kfree(info->stripe_hash_table);
+ info->stripe_hash_table = NULL;
+}
+
+/*
+ * insert an rbio into the stripe cache. It
+ * must have already been prepared by calling
+ * cache_rbio_pages
+ *
+ * If this rbio was already cached, it gets
+ * moved to the front of the lru.
+ *
+ * If the size of the rbio cache is too big, we
+ * prune an item.
+ */
+static void cache_rbio(struct btrfs_raid_bio *rbio)
+{
+ struct btrfs_stripe_hash_table *table;
+ unsigned long flags;
+
+ if (!test_bit(RBIO_CACHE_READY_BIT, &rbio->flags))
+ return;
+
+ table = rbio->fs_info->stripe_hash_table;
+
+ spin_lock_irqsave(&table->cache_lock, flags);
+ spin_lock(&rbio->bio_list_lock);
+
+ /* 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);
+
+ if (!list_empty(&rbio->stripe_cache)){
+ list_move(&rbio->stripe_cache, &table->stripe_cache);
+ } else {
+ list_add(&rbio->stripe_cache, &table->stripe_cache);
+ table->cache_size += 1;
+ }
+
+ spin_unlock(&rbio->bio_list_lock);
+
+ if (table->cache_size > RBIO_CACHE_SIZE) {
+ struct btrfs_raid_bio *found;
+
+ found = list_entry(table->stripe_cache.prev,
+ struct btrfs_raid_bio,
+ stripe_cache);
+
+ if (found != rbio)
+ __remove_rbio_from_cache(found);
+ }
+
+ spin_unlock_irqrestore(&table->cache_lock, flags);
+ return;
+}
+
+/*
+ * helper function to run the xor_blocks api. It is only
+ * able to do MAX_XOR_BLOCKS at a time, so we need to
+ * loop through.
+ */
+static void run_xor(void **pages, int src_cnt, ssize_t len)
+{
+ int src_off = 0;
+ int xor_src_cnt = 0;
+ void *dest = pages[src_cnt];
+
+ while(src_cnt > 0) {
+ xor_src_cnt = min(src_cnt, MAX_XOR_BLOCKS);
+ xor_blocks(xor_src_cnt, len, dest, pages + src_off);
+
+ src_cnt -= xor_src_cnt;
+ src_off += xor_src_cnt;
+ }
+}
+
+/*
+ * returns true if the bio list inside this rbio
+ * covers an entire stripe (no rmw required).
+ * Must be called with the bio list lock held, or
+ * at a time when you know it is impossible to add
+ * new bios into the list
+ */
+static int __rbio_is_full(struct btrfs_raid_bio *rbio)
+{
+ unsigned long size = rbio->bio_list_bytes;
+ int ret = 1;
+
+ if (size != rbio->nr_data * rbio->stripe_len)
+ ret = 0;
+
+ BUG_ON(size > rbio->nr_data * rbio->stripe_len);
+ return ret;
+}
+
+static int rbio_is_full(struct btrfs_raid_bio *rbio)
+{
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&rbio->bio_list_lock, flags);
+ ret = __rbio_is_full(rbio);
+ spin_unlock_irqrestore(&rbio->bio_list_lock, flags);
+ return ret;
+}
+
+/*
+ * returns 1 if it is safe to merge two rbios together.
+ * The merging is safe if the two rbios correspond to
+ * the same stripe and if they are both going in the same
+ * direction (read vs write), and if neither one is
+ * locked for final IO
+ *
+ * The caller is responsible for locking such that
+ * rmw_locked is safe to test
+ */
+static int rbio_can_merge(struct btrfs_raid_bio *last,
+ struct btrfs_raid_bio *cur)
+{
+ if (test_bit(RBIO_RMW_LOCKED_BIT, &last->flags) ||
+ test_bit(RBIO_RMW_LOCKED_BIT, &cur->flags))
+ return 0;
+
+ /*
+ * we can't merge with cached rbios, since the
+ * idea is that when we merge the destination
+ * rbio is going to run our IO for us. We can
+ * steal from cached rbio's though, other functions
+ * handle that.
+ */
+ if (test_bit(RBIO_CACHE_BIT, &last->flags) ||
+ test_bit(RBIO_CACHE_BIT, &cur->flags))
+ return 0;
+
+ if (last->raid_map[0] !=
+ cur->raid_map[0])
+ return 0;
+
+ /* reads can't merge with writes */
+ if (last->read_rebuild !=
+ cur->read_rebuild) {
+ return 0;
+ }
+
+ return 1;
+}
+
+/*
+ * helper to index into the pstripe
+ */
+static struct page *rbio_pstripe_page(struct btrfs_raid_bio *rbio, int index)
+{
+ index += (rbio->nr_data * rbio->stripe_len) >> PAGE_CACHE_SHIFT;
+ return rbio->stripe_pages[index];
+}
+
+/*
+ * helper to index into the qstripe, returns null
+ * if there is no qstripe
+ */
+static struct page *rbio_qstripe_page(struct btrfs_raid_bio *rbio, int index)
+{
+ if (rbio->nr_data + 1 == rbio->bbio->num_stripes)
+ return NULL;
+
+ index += ((rbio->nr_data + 1) * rbio->stripe_len) >>
+ PAGE_CACHE_SHIFT;
+ return rbio->stripe_pages[index];
+}
+
+/*
+ * The first stripe in the table for a logical address
+ * has the lock. rbios are added in one of three ways:
+ *
+ * 1) Nobody has the stripe locked yet. The rbio is given
+ * the lock and 0 is returned. The caller must start the IO
+ * themselves.
+ *
+ * 2) Someone has the stripe locked, but we're able to merge
+ * with the lock owner. The rbio is freed and the IO will
+ * start automatically along with the existing rbio. 1 is returned.
+ *
+ * 3) Someone has the stripe locked, but we're not able to merge.
+ * The rbio is added to the lock owner's plug list, or merged into
+ * an rbio already on the plug list. When the lock owner unlocks,
+ * the next rbio on the list is run and the IO is started automatically.
+ * 1 is returned
+ *
+ * If we return 0, the caller still owns the rbio and must continue with
+ * IO submission. If we return 1, the caller must assume the rbio has
+ * already been freed.
+ */
+static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
+{
+ int bucket = rbio_bucket(rbio);
+ struct btrfs_stripe_hash *h = rbio->fs_info->stripe_hash_table->table + bucket;
+ struct btrfs_raid_bio *cur;
+ struct btrfs_raid_bio *pending;
+ unsigned long flags;
+ DEFINE_WAIT(wait);
+ 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->raid_map[0] == rbio->raid_map[0]) {
+ spin_lock(&cur->bio_list_lock);
+
+ /* can we steal this cached rbio's pages? */
+ if (bio_list_empty(&cur->bio_list) &&
+ list_empty(&cur->plug_list) &&
+ 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);
+
+ steal_rbio(cur, rbio);
+ cache_drop = cur;
+ spin_unlock(&cur->bio_list_lock);
+
+ goto lockit;
+ }
+
+ /* can we merge into the lock owner? */
+ if (rbio_can_merge(cur, rbio)) {
+ merge_rbio(cur, rbio);
+ spin_unlock(&cur->bio_list_lock);
+ freeit = rbio;
+ ret = 1;
+ goto out;
+ }
+
+
+ /*
+ * we couldn't merge with the running
+ * rbio, see if we can merge with the
+ * pending ones. We don't have to
+ * check for rmw_locked because there
+ * is no way they are inside finish_rmw
+ * right now
+ */
+ list_for_each_entry(pending, &cur->plug_list,
+ plug_list) {
+ if (rbio_can_merge(pending, rbio)) {
+ merge_rbio(pending, rbio);
+ spin_unlock(&cur->bio_list_lock);
+ freeit = rbio;
+ ret = 1;
+ goto out;
+ }
+ }
+
+ /* no merging, put us on the tail of the plug list,
+ * our rbio will be started with the currently
+ * running rbio unlocks
+ */
+ list_add_tail(&rbio->plug_list, &cur->plug_list);
+ spin_unlock(&cur->bio_list_lock);
+ ret = 1;
+ goto out;
+ }
+ }
+lockit:
+ atomic_inc(&rbio->refs);
+ list_add(&rbio->hash_list, &h->hash_list);
+out:
+ spin_unlock_irqrestore(&h->lock, flags);
+ if (cache_drop)
+ remove_rbio_from_cache(cache_drop);
+ if (freeit)
+ __free_raid_bio(freeit);
+ return ret;
+}
+
+/*
+ * called as rmw or parity rebuild is completed. If the plug list has more
+ * rbios waiting for this stripe, the next one on the list will be started
+ */
+static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
+{
+ int bucket;
+ struct btrfs_stripe_hash *h;
+ unsigned long flags;
+ int keep_cache = 0;
+
+ bucket = rbio_bucket(rbio);
+ h = rbio->fs_info->stripe_hash_table->table + bucket;
+
+ if (list_empty(&rbio->plug_list))
+ cache_rbio(rbio);
+
+ spin_lock_irqsave(&h->lock, flags);
+ spin_lock(&rbio->bio_list_lock);
+
+ if (!list_empty(&rbio->hash_list)) {
+ /*
+ * if we're still cached and there is no other IO
+ * to perform, just leave this rbio here for others
+ * to steal from later
+ */
+ if (list_empty(&rbio->plug_list) &&
+ test_bit(RBIO_CACHE_BIT, &rbio->flags)) {
+ keep_cache = 1;
+ clear_bit(RBIO_RMW_LOCKED_BIT, &rbio->flags);
+ BUG_ON(!bio_list_empty(&rbio->bio_list));
+ goto done;
+ }
+
+ list_del_init(&rbio->hash_list);
+ atomic_dec(&rbio->refs);
+
+ /*
+ * we use the plug list to hold all the rbios
+ * waiting for the chance to lock this stripe.
+ * hand the lock over to one of them.
+ */
+ if (!list_empty(&rbio->plug_list)) {
+ struct btrfs_raid_bio *next;
+ struct list_head *head = rbio->plug_list.next;
+
+ next = list_entry(head, struct btrfs_raid_bio,
+ plug_list);
+
+ list_del_init(&rbio->plug_list);
+
+ list_add(&next->hash_list, &h->hash_list);
+ atomic_inc(&next->refs);
+ spin_unlock(&rbio->bio_list_lock);
+ spin_unlock_irqrestore(&h->lock, flags);
+
+ if (next->read_rebuild)
+ async_read_rebuild(next);
+ else {
+ steal_rbio(rbio, next);
+ async_rmw_stripe(next);
+ }
+
+ goto done_nolock;
+ } else if (waitqueue_active(&h->wait)) {
+ spin_unlock(&rbio->bio_list_lock);
+ spin_unlock_irqrestore(&h->lock, flags);
+ wake_up(&h->wait);
+ goto done_nolock;
+ }
+ }
+done:
+ spin_unlock(&rbio->bio_list_lock);
+ spin_unlock_irqrestore(&h->lock, flags);
+
+done_nolock:
+ if (!keep_cache)
+ remove_rbio_from_cache(rbio);
+}
+
+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))
+ return;
+
+ WARN_ON(!list_empty(&rbio->stripe_cache));
+ WARN_ON(!list_empty(&rbio->hash_list));
+ WARN_ON(!bio_list_empty(&rbio->bio_list));
+
+ for (i = 0; i < rbio->nr_pages; i++) {
+ if (rbio->stripe_pages[i]) {
+ __free_page(rbio->stripe_pages[i]);
+ rbio->stripe_pages[i] = NULL;
+ }
+ }
+ kfree(rbio->raid_map);
+ kfree(rbio->bbio);
+ kfree(rbio);
+}
+
+static void free_raid_bio(struct btrfs_raid_bio *rbio)
+{
+ unlock_stripe(rbio);
+ __free_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, int uptodate)
+{
+ struct bio *cur = bio_list_get(&rbio->bio_list);
+ struct bio *next;
+ free_raid_bio(rbio);
+
+ while (cur) {
+ next = cur->bi_next;
+ cur->bi_next = NULL;
+ if (uptodate)
+ set_bit(BIO_UPTODATE, &cur->bi_flags);
+ bio_endio(cur, err);
+ cur = next;
+ }
+}
+
+/*
+ * end io function used by finish_rmw. When we finally
+ * get here, we've written a full stripe
+ */
+static void raid_write_end_io(struct bio *bio, int err)
+{
+ struct btrfs_raid_bio *rbio = bio->bi_private;
+
+ if (err)
+ fail_bio_stripe(rbio, bio);
+
+ bio_put(bio);
+
+ if (!atomic_dec_and_test(&rbio->bbio->stripes_pending))
+ return;
+
+ err = 0;
+
+ /* OK, we have read all the stripes we need to. */
+ if (atomic_read(&rbio->bbio->error) > rbio->bbio->max_errors)
+ err = -EIO;
+
+ rbio_orig_end_io(rbio, err, 0);
+ return;
+}
+
+/*
+ * the read/modify/write code wants to use the original bio for
+ * any pages it included, and then use the rbio for everything
+ * else. This function decides if a given index (stripe number)
+ * and page number in that stripe fall inside the original bio
+ * or the rbio.
+ *
+ * if you set bio_list_only, you'll get a NULL back for any ranges
+ * that are outside the bio_list
+ *
+ * This doesn't take any refs on anything, you get a bare page pointer
+ * and the caller must bump refs as required.
+ *
+ * You must call index_rbio_pages once before you can trust
+ * the answers from this function.
+ */
+static struct page *page_in_rbio(struct btrfs_raid_bio *rbio,
+ int index, int pagenr, int bio_list_only)
+{
+ int chunk_page;
+ struct page *p = NULL;
+
+ chunk_page = index * (rbio->stripe_len >> PAGE_SHIFT) + pagenr;
+
+ spin_lock_irq(&rbio->bio_list_lock);
+ p = rbio->bio_pages[chunk_page];
+ spin_unlock_irq(&rbio->bio_list_lock);
+
+ if (p || bio_list_only)
+ return p;
+
+ return rbio->stripe_pages[chunk_page];
+}
+
+/*
+ * number of pages we need for the entire stripe across all the
+ * drives
+ */
+static unsigned long rbio_nr_pages(unsigned long stripe_len, int nr_stripes)
+{
+ unsigned long nr = stripe_len * nr_stripes;
+ return (nr + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+}
+
+/*
+ * allocation and initial setup for the btrfs_raid_bio. Not
+ * this does not allocate any pages for rbio->pages.
+ */
+static struct btrfs_raid_bio *alloc_rbio(struct btrfs_root *root,
+ struct btrfs_bio *bbio, u64 *raid_map,
+ u64 stripe_len)
+{
+ struct btrfs_raid_bio *rbio;
+ int nr_data = 0;
+ int num_pages = rbio_nr_pages(stripe_len, bbio->num_stripes);
+ void *p;
+
+ rbio = kzalloc(sizeof(*rbio) + num_pages * sizeof(struct page *) * 2,
+ GFP_NOFS);
+ if (!rbio) {
+ kfree(raid_map);
+ kfree(bbio);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ bio_list_init(&rbio->bio_list);
+ INIT_LIST_HEAD(&rbio->plug_list);
+ spin_lock_init(&rbio->bio_list_lock);
+ INIT_LIST_HEAD(&rbio->stripe_cache);
+ INIT_LIST_HEAD(&rbio->hash_list);
+ rbio->bbio = bbio;
+ rbio->raid_map = raid_map;
+ rbio->fs_info = root->fs_info;
+ rbio->stripe_len = stripe_len;
+ rbio->nr_pages = num_pages;
+ rbio->faila = -1;
+ rbio->failb = -1;
+ atomic_set(&rbio->refs, 1);
+
+ /*
+ * the stripe_pages and bio_pages array point to the extra
+ * memory we allocated past the end of the rbio
+ */
+ p = rbio + 1;
+ rbio->stripe_pages = p;
+ rbio->bio_pages = p + sizeof(struct page *) * num_pages;
+
+ if (raid_map[bbio->num_stripes - 1] == RAID6_Q_STRIPE)
+ nr_data = bbio->num_stripes - 2;
+ else
+ nr_data = bbio->num_stripes - 1;
+
+ rbio->nr_data = nr_data;
+ return rbio;
+}
+
+/* allocate pages for all the stripes in the bio, including parity */
+static int alloc_rbio_pages(struct btrfs_raid_bio *rbio)
+{
+ int i;
+ struct page *page;
+
+ for (i = 0; i < rbio->nr_pages; i++) {
+ if (rbio->stripe_pages[i])
+ continue;
+ page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+ if (!page)
+ return -ENOMEM;
+ rbio->stripe_pages[i] = page;
+ ClearPageUptodate(page);
+ }
+ return 0;
+}
+
+/* allocate pages for just the p/q stripes */
+static int alloc_rbio_parity_pages(struct btrfs_raid_bio *rbio)
+{
+ int i;
+ struct page *page;
+
+ i = (rbio->nr_data * rbio->stripe_len) >> PAGE_CACHE_SHIFT;
+
+ for (; i < rbio->nr_pages; i++) {
+ if (rbio->stripe_pages[i])
+ continue;
+ page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+ if (!page)
+ return -ENOMEM;
+ rbio->stripe_pages[i] = page;
+ }
+ return 0;
+}
+
+/*
+ * add a single page from a specific stripe into our list of bios for IO
+ * this will try to merge into existing bios if possible, and returns
+ * zero if all went well.
+ */
+static int rbio_add_io_page(struct btrfs_raid_bio *rbio,
+ struct bio_list *bio_list,
+ struct page *page,
+ int stripe_nr,
+ unsigned long page_index,
+ unsigned long bio_max_len)
+{
+ struct bio *last = bio_list->tail;
+ u64 last_end = 0;
+ int ret;
+ struct bio *bio;
+ struct btrfs_bio_stripe *stripe;
+ u64 disk_start;
+
+ stripe = &rbio->bbio->stripes[stripe_nr];
+ disk_start = stripe->physical + (page_index << PAGE_CACHE_SHIFT);
+
+ /* if the device is missing, just fail this stripe */
+ if (!stripe->dev->bdev)
+ return fail_rbio_index(rbio, stripe_nr);
+
+ /* see if we can add this page onto our existing bio */
+ if (last) {
+ last_end = (u64)last->bi_sector << 9;
+ last_end += last->bi_size;
+
+ /*
+ * we can't merge these if they are from different
+ * devices or if they are not contiguous
+ */
+ if (last_end == disk_start && stripe->dev->bdev &&
+ test_bit(BIO_UPTODATE, &last->bi_flags) &&
+ last->bi_bdev == stripe->dev->bdev) {
+ ret = bio_add_page(last, page, PAGE_CACHE_SIZE, 0);
+ if (ret == PAGE_CACHE_SIZE)
+ return 0;
+ }
+ }
+
+ /* put a new bio on the list */
+ bio = bio_alloc(GFP_NOFS, bio_max_len >> PAGE_SHIFT?:1);
+ if (!bio)
+ return -ENOMEM;
+
+ bio->bi_size = 0;
+ bio->bi_bdev = stripe->dev->bdev;
+ bio->bi_sector = disk_start >> 9;
+ set_bit(BIO_UPTODATE, &bio->bi_flags);
+
+ bio_add_page(bio, page, PAGE_CACHE_SIZE, 0);
+ bio_list_add(bio_list, bio);
+ return 0;
+}
+
+/*
+ * while we're doing the read/modify/write cycle, we could
+ * have errors in reading pages off the disk. This checks
+ * for errors and if we're not able to read the page it'll
+ * trigger parity reconstruction. The rmw will be finished
+ * after we've reconstructed the failed stripes
+ */
+static void validate_rbio_for_rmw(struct btrfs_raid_bio *rbio)
+{
+ if (rbio->faila >= 0 || rbio->failb >= 0) {
+ BUG_ON(rbio->faila == rbio->bbio->num_stripes - 1);
+ __raid56_parity_recover(rbio);
+ } else {
+ finish_rmw(rbio);
+ }
+}
+
+/*
+ * these are just the pages from the rbio array, not from anything
+ * the FS sent down to us
+ */
+static struct page *rbio_stripe_page(struct btrfs_raid_bio *rbio, int stripe, int page)
+{
+ int index;
+ index = stripe * (rbio->stripe_len >> PAGE_CACHE_SHIFT);
+ index += page;
+ return rbio->stripe_pages[index];
+}
+
+/*
+ * helper function to walk our bio list and populate the bio_pages array with
+ * the result. This seems expensive, but it is faster than constantly
+ * searching through the bio list as we setup the IO in finish_rmw or stripe
+ * reconstruction.
+ *
+ * This must be called before you trust the answers from page_in_rbio
+ */
+static void index_rbio_pages(struct btrfs_raid_bio *rbio)
+{
+ struct bio *bio;
+ u64 start;
+ unsigned long stripe_offset;
+ unsigned long page_index;
+ struct page *p;
+ int i;
+
+ spin_lock_irq(&rbio->bio_list_lock);
+ bio_list_for_each(bio, &rbio->bio_list) {
+ start = (u64)bio->bi_sector << 9;
+ stripe_offset = start - rbio->raid_map[0];
+ page_index = stripe_offset >> PAGE_CACHE_SHIFT;
+
+ for (i = 0; i < bio->bi_vcnt; i++) {
+ p = bio->bi_io_vec[i].bv_page;
+ rbio->bio_pages[page_index + i] = p;
+ }
+ }
+ spin_unlock_irq(&rbio->bio_list_lock);
+}
+
+/*
+ * this is called from one of two situations. We either
+ * have a full stripe from the higher layers, or we've read all
+ * the missing bits off disk.
+ *
+ * This will calculate the parity and then send down any
+ * changed blocks.
+ */
+static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
+{
+ struct btrfs_bio *bbio = rbio->bbio;
+ void *pointers[bbio->num_stripes];
+ int stripe_len = rbio->stripe_len;
+ int nr_data = rbio->nr_data;
+ int stripe;
+ int pagenr;
+ int p_stripe = -1;
+ int q_stripe = -1;
+ struct bio_list bio_list;
+ struct bio *bio;
+ int pages_per_stripe = stripe_len >> PAGE_CACHE_SHIFT;
+ int ret;
+
+ bio_list_init(&bio_list);
+
+ if (bbio->num_stripes - rbio->nr_data == 1) {
+ p_stripe = bbio->num_stripes - 1;
+ } else if (bbio->num_stripes - rbio->nr_data == 2) {
+ p_stripe = bbio->num_stripes - 2;
+ q_stripe = bbio->num_stripes - 1;
+ } else {
+ BUG();
+ }
+
+ /* at this point we either have a full stripe,
+ * or we've read the full stripe from the drive.
+ * recalculate the parity and write the new results.
+ *
+ * We're not allowed to add any new bios to the
+ * bio list here, anyone else that wants to
+ * change this stripe needs to do their own rmw.
+ */
+ spin_lock_irq(&rbio->bio_list_lock);
+ set_bit(RBIO_RMW_LOCKED_BIT, &rbio->flags);
+ spin_unlock_irq(&rbio->bio_list_lock);
+
+ atomic_set(&rbio->bbio->error, 0);
+
+ /*
+ * now that we've set rmw_locked, run through the
+ * bio list one last time and map the page pointers
+ *
+ * We don't cache full rbios because we're assuming
+ * the higher layers are unlikely to use this area of
+ * the disk again soon. If they do use it again,
+ * hopefully they will send another full bio.
+ */
+ index_rbio_pages(rbio);
+ if (!rbio_is_full(rbio))
+ cache_rbio_pages(rbio);
+ else
+ clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
+
+ for (pagenr = 0; pagenr < pages_per_stripe; pagenr++) {
+ struct page *p;
+ /* first collect one page from each data stripe */
+ for (stripe = 0; stripe < nr_data; stripe++) {
+ p = page_in_rbio(rbio, stripe, pagenr, 0);
+ pointers[stripe] = kmap(p);
+ }
+
+ /* then add the parity stripe */
+ p = rbio_pstripe_page(rbio, pagenr);
+ SetPageUptodate(p);
+ pointers[stripe++] = kmap(p);
+
+ if (q_stripe != -1) {
+
+ /*
+ * raid6, add the qstripe and call the
+ * library function to fill in our p/q
+ */
+ p = rbio_qstripe_page(rbio, pagenr);
+ SetPageUptodate(p);
+ pointers[stripe++] = kmap(p);
+
+ raid6_call.gen_syndrome(bbio->num_stripes, PAGE_SIZE,
+ pointers);
+ } else {
+ /* raid5 */
+ memcpy(pointers[nr_data], pointers[0], PAGE_SIZE);
+ run_xor(pointers + 1, nr_data - 1, PAGE_CACHE_SIZE);
+ }
+
+
+ for (stripe = 0; stripe < bbio->num_stripes; stripe++)
+ kunmap(page_in_rbio(rbio, stripe, pagenr, 0));
+ }
+
+ /*
+ * time to start writing. Make bios for everything from the
+ * higher layers (the bio_list in our rbio) and our p/q. Ignore
+ * everything else.
+ */
+ for (stripe = 0; stripe < bbio->num_stripes; stripe++) {
+ for (pagenr = 0; pagenr < pages_per_stripe; pagenr++) {
+ struct page *page;
+ if (stripe < rbio->nr_data) {
+ page = page_in_rbio(rbio, stripe, pagenr, 1);
+ if (!page)
+ continue;
+ } else {
+ page = rbio_stripe_page(rbio, stripe, pagenr);
+ }
+
+ ret = rbio_add_io_page(rbio, &bio_list,
+ page, stripe, pagenr, rbio->stripe_len);
+ if (ret)
+ goto cleanup;
+ }
+ }
+
+ atomic_set(&bbio->stripes_pending, bio_list_size(&bio_list));
+ BUG_ON(atomic_read(&bbio->stripes_pending) == 0);
+
+ while (1) {
+ bio = bio_list_pop(&bio_list);
+ if (!bio)
+ break;
+
+ bio->bi_private = rbio;
+ bio->bi_end_io = raid_write_end_io;
+ BUG_ON(!test_bit(BIO_UPTODATE, &bio->bi_flags));
+ submit_bio(WRITE, bio);
+ }
+ return;
+
+cleanup:
+ rbio_orig_end_io(rbio, -EIO, 0);
+}
+
+/*
+ * helper to find the stripe number for a given bio. Used to figure out which
+ * stripe has failed. This expects the bio to correspond to a physical disk,
+ * so it looks up based on physical sector numbers.
+ */
+static int find_bio_stripe(struct btrfs_raid_bio *rbio,
+ struct bio *bio)
+{
+ u64 physical = bio->bi_sector;
+ u64 stripe_start;
+ int i;
+ struct btrfs_bio_stripe *stripe;
+
+ physical <<= 9;
+
+ for (i = 0; i < rbio->bbio->num_stripes; i++) {
+ stripe = &rbio->bbio->stripes[i];
+ stripe_start = stripe->physical;
+ if (physical >= stripe_start &&
+ physical < stripe_start + rbio->stripe_len) {
+ return i;
+ }
+ }
+ return -1;
+}
+
+/*
+ * helper to find the stripe number for a given
+ * bio (before mapping). Used to figure out which stripe has
+ * failed. This looks up based on logical block numbers.
+ */
+static int find_logical_bio_stripe(struct btrfs_raid_bio *rbio,
+ struct bio *bio)
+{
+ u64 logical = bio->bi_sector;
+ u64 stripe_start;
+ int i;
+
+ logical <<= 9;
+
+ for (i = 0; i < rbio->nr_data; i++) {
+ stripe_start = rbio->raid_map[i];
+ if (logical >= stripe_start &&
+ logical < stripe_start + rbio->stripe_len) {
+ return i;
+ }
+ }
+ return -1;
+}
+
+/*
+ * returns -EIO if we had too many failures
+ */
+static int fail_rbio_index(struct btrfs_raid_bio *rbio, int failed)
+{
+ unsigned long flags;
+ int ret = 0;
+
+ spin_lock_irqsave(&rbio->bio_list_lock, flags);
+
+ /* we already know this stripe is bad, move on */
+ if (rbio->faila == failed || rbio->failb == failed)
+ goto out;
+
+ if (rbio->faila == -1) {
+ /* first failure on this rbio */
+ rbio->faila = failed;
+ atomic_inc(&rbio->bbio->error);
+ } else if (rbio->failb == -1) {
+ /* second failure on this rbio */
+ rbio->failb = failed;
+ atomic_inc(&rbio->bbio->error);
+ } else {
+ ret = -EIO;
+ }
+out:
+ spin_unlock_irqrestore(&rbio->bio_list_lock, flags);
+
+ return ret;
+}
+
+/*
+ * helper to fail a stripe based on a physical disk
+ * bio.
+ */
+static int fail_bio_stripe(struct btrfs_raid_bio *rbio,
+ struct bio *bio)
+{
+ int failed = find_bio_stripe(rbio, bio);
+
+ if (failed < 0)
+ return -EIO;
+
+ return fail_rbio_index(rbio, failed);
+}
+
+/*
+ * this sets each page in the bio uptodate. It should only be used on private
+ * rbio pages, nothing that comes in from the higher layers
+ */
+static void set_bio_pages_uptodate(struct bio *bio)
+{
+ int i;
+ struct page *p;
+
+ for (i = 0; i < bio->bi_vcnt; i++) {
+ p = bio->bi_io_vec[i].bv_page;
+ SetPageUptodate(p);
+ }
+}
+
+/*
+ * end io for the read phase of the rmw cycle. All the bios here are physical
+ * stripe bios we've read from the disk so we can recalculate the parity of the
+ * stripe.
+ *
+ * This will usually kick off finish_rmw once all the bios are read in, but it
+ * may trigger parity reconstruction if we had any errors along the way
+ */
+static void raid_rmw_end_io(struct bio *bio, int err)
+{
+ struct btrfs_raid_bio *rbio = bio->bi_private;
+
+ if (err)
+ fail_bio_stripe(rbio, bio);
+ else
+ set_bio_pages_uptodate(bio);
+
+ bio_put(bio);
+
+ if (!atomic_dec_and_test(&rbio->bbio->stripes_pending))
+ return;
+
+ err = 0;
+ if (atomic_read(&rbio->bbio->error) > rbio->bbio->max_errors)
+ goto cleanup;
+
+ /*
+ * this will normally call finish_rmw to start our write
+ * but if there are any failed stripes we'll reconstruct
+ * from parity first
+ */
+ validate_rbio_for_rmw(rbio);
+ return;
+
+cleanup:
+
+ rbio_orig_end_io(rbio, -EIO, 0);
+}
+
+static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
+{
+ rbio->work.flags = 0;
+ rbio->work.func = rmw_work;
+
+ btrfs_queue_worker(&rbio->fs_info->rmw_workers,
+ &rbio->work);
+}
+
+static void async_read_rebuild(struct btrfs_raid_bio *rbio)
+{
+ rbio->work.flags = 0;
+ rbio->work.func = read_rebuild_work;
+
+ btrfs_queue_worker(&rbio->fs_info->rmw_workers,
+ &rbio->work);
+}
+
+/*
+ * the stripe must be locked by the caller. It will
+ * unlock after all the writes are done
+ */
+static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
+{
+ int bios_to_read = 0;
+ struct btrfs_bio *bbio = rbio->bbio;
+ struct bio_list bio_list;
+ int ret;
+ int nr_pages = (rbio->stripe_len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+ int pagenr;
+ int stripe;
+ struct bio *bio;
+
+ bio_list_init(&bio_list);
+
+ ret = alloc_rbio_pages(rbio);
+ if (ret)
+ goto cleanup;
+
+ index_rbio_pages(rbio);
+
+ atomic_set(&rbio->bbio->error, 0);
+ /*
+ * build a list of bios to read all the missing parts of this
+ * stripe
+ */
+ for (stripe = 0; stripe < rbio->nr_data; stripe++) {
+ for (pagenr = 0; pagenr < nr_pages; pagenr++) {
+ struct page *page;
+ /*
+ * we want to find all the pages missing from
+ * the rbio and read them from the disk. If
+ * page_in_rbio finds a page in the bio list
+ * we don't need to read it off the stripe.
+ */
+ page = page_in_rbio(rbio, stripe, pagenr, 1);
+ if (page)
+ continue;
+
+ page = rbio_stripe_page(rbio, stripe, pagenr);
+ /*
+ * the bio cache may have handed us an uptodate
+ * page. If so, be happy and use it
+ */
+ if (PageUptodate(page))
+ continue;
+
+ ret = rbio_add_io_page(rbio, &bio_list, page,
+ stripe, pagenr, rbio->stripe_len);
+ if (ret)
+ goto cleanup;
+ }
+ }
+
+ bios_to_read = bio_list_size(&bio_list);
+ if (!bios_to_read) {
+ /*
+ * this can happen if others have merged with
+ * us, it means there is nothing left to read.
+ * But if there are missing devices it may not be
+ * safe to do the full stripe write yet.
+ */
+ goto finish;
+ }
+
+ /*
+ * the bbio may be freed once we submit the last bio. Make sure
+ * not to touch it after that
+ */
+ atomic_set(&bbio->stripes_pending, bios_to_read);
+ while (1) {
+ bio = bio_list_pop(&bio_list);
+ if (!bio)
+ break;
+
+ bio->bi_private = rbio;
+ bio->bi_end_io = raid_rmw_end_io;
+
+ btrfs_bio_wq_end_io(rbio->fs_info, bio,
+ BTRFS_WQ_ENDIO_RAID56);
+
+ BUG_ON(!test_bit(BIO_UPTODATE, &bio->bi_flags));
+ submit_bio(READ, bio);
+ }
+ /* the actual write will happen once the reads are done */
+ return 0;
+
+cleanup:
+ rbio_orig_end_io(rbio, -EIO, 0);
+ return -EIO;
+
+finish:
+ validate_rbio_for_rmw(rbio);
+ return 0;
+}
+
+/*
+ * if the upper layers pass in a full stripe, we thank them by only allocating
+ * enough pages to hold the parity, and sending it all down quickly.
+ */
+static int full_stripe_write(struct btrfs_raid_bio *rbio)
+{
+ int ret;
+
+ ret = alloc_rbio_parity_pages(rbio);
+ if (ret)
+ return ret;
+
+ ret = lock_stripe_add(rbio);
+ if (ret == 0)
+ finish_rmw(rbio);
+ return 0;
+}
+
+/*
+ * partial stripe writes get handed over to async helpers.
+ * We're really hoping to merge a few more writes into this
+ * rbio before calculating new parity
+ */
+static int partial_stripe_write(struct btrfs_raid_bio *rbio)
+{
+ int ret;
+
+ ret = lock_stripe_add(rbio);
+ if (ret == 0)
+ async_rmw_stripe(rbio);
+ return 0;
+}
+
+/*
+ * sometimes while we were reading from the drive to
+ * recalculate parity, enough new bios come into create
+ * a full stripe. So we do a check here to see if we can
+ * go directly to finish_rmw
+ */
+static int __raid56_parity_write(struct btrfs_raid_bio *rbio)
+{
+ /* head off into rmw land if we don't have a full stripe */
+ if (!rbio_is_full(rbio))
+ return partial_stripe_write(rbio);
+ return full_stripe_write(rbio);
+}
+
+/*
+ * We use plugging call backs to collect full stripes.
+ * Any time we get a partial stripe write while plugged
+ * we collect it into a list. When the unplug comes down,
+ * we sort the list by logical block number and merge
+ * everything we can into the same rbios
+ */
+struct btrfs_plug_cb {
+ struct blk_plug_cb cb;
+ struct btrfs_fs_info *info;
+ struct list_head rbio_list;
+ struct btrfs_work work;
+};
+
+/*
+ * rbios on the plug list are sorted for easier merging.
+ */
+static int plug_cmp(void *priv, struct list_head *a, struct list_head *b)
+{
+ struct btrfs_raid_bio *ra = container_of(a, struct btrfs_raid_bio,
+ plug_list);
+ struct btrfs_raid_bio *rb = container_of(b, struct btrfs_raid_bio,
+ plug_list);
+ u64 a_sector = ra->bio_list.head->bi_sector;
+ u64 b_sector = rb->bio_list.head->bi_sector;
+
+ if (a_sector < b_sector)
+ return -1;
+ if (a_sector > b_sector)
+ return 1;
+ return 0;
+}
+
+static void run_plug(struct btrfs_plug_cb *plug)
+{
+ struct btrfs_raid_bio *cur;
+ struct btrfs_raid_bio *last = NULL;
+
+ /*
+ * sort our plug list then try to merge
+ * everything we can in hopes of creating full
+ * stripes.
+ */
+ list_sort(NULL, &plug->rbio_list, plug_cmp);
+ while (!list_empty(&plug->rbio_list)) {
+ cur = list_entry(plug->rbio_list.next,
+ struct btrfs_raid_bio, plug_list);
+ list_del_init(&cur->plug_list);
+
+ if (rbio_is_full(cur)) {
+ /* we have a full stripe, send it down */
+ full_stripe_write(cur);
+ continue;
+ }
+ if (last) {
+ if (rbio_can_merge(last, cur)) {
+ merge_rbio(last, cur);
+ __free_raid_bio(cur);
+ continue;
+
+ }
+ __raid56_parity_write(last);
+ }
+ last = cur;
+ }
+ if (last) {
+ __raid56_parity_write(last);
+ }
+ kfree(plug);
+}
+
+/*
+ * if the unplug comes from schedule, we have to push the
+ * work off to a helper thread
+ */
+static void unplug_work(struct btrfs_work *work)
+{
+ struct btrfs_plug_cb *plug;
+ plug = container_of(work, struct btrfs_plug_cb, work);
+ run_plug(plug);
+}
+
+static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
+{
+ struct btrfs_plug_cb *plug;
+ plug = container_of(cb, struct btrfs_plug_cb, cb);
+
+ if (from_schedule) {
+ plug->work.flags = 0;
+ plug->work.func = unplug_work;
+ btrfs_queue_worker(&plug->info->rmw_workers,
+ &plug->work);
+ return;
+ }
+ run_plug(plug);
+}
+
+/*
+ * our main entry point for writes from the rest of the FS.
+ */
+int raid56_parity_write(struct btrfs_root *root, struct bio *bio,
+ struct btrfs_bio *bbio, u64 *raid_map,
+ u64 stripe_len)
+{
+ struct btrfs_raid_bio *rbio;
+ struct btrfs_plug_cb *plug = NULL;
+ struct blk_plug_cb *cb;
+
+ rbio = alloc_rbio(root, bbio, raid_map, stripe_len);
+ if (IS_ERR(rbio)) {
+ kfree(raid_map);
+ kfree(bbio);
+ return PTR_ERR(rbio);
+ }
+ bio_list_add(&rbio->bio_list, bio);
+ rbio->bio_list_bytes = bio->bi_size;
+
+ /*
+ * don't plug on full rbios, just get them out the door
+ * as quickly as we can
+ */
+ if (rbio_is_full(rbio))
+ return full_stripe_write(rbio);
+
+ cb = blk_check_plugged(btrfs_raid_unplug, root->fs_info,
+ sizeof(*plug));
+ if (cb) {
+ plug = container_of(cb, struct btrfs_plug_cb, cb);
+ if (!plug->info) {
+ plug->info = root->fs_info;
+ INIT_LIST_HEAD(&plug->rbio_list);
+ }
+ list_add_tail(&rbio->plug_list, &plug->rbio_list);
+ } else {
+ return __raid56_parity_write(rbio);
+ }
+ return 0;
+}
+
+/*
+ * all parity reconstruction happens here. We've read in everything
+ * we can find from the drives and this does the heavy lifting of
+ * sorting the good from the bad.
+ */
+static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
+{
+ int pagenr, stripe;
+ void **pointers;
+ int faila = -1, failb = -1;
+ int nr_pages = (rbio->stripe_len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+ struct page *page;
+ int err;
+ int i;
+
+ pointers = kzalloc(rbio->bbio->num_stripes * sizeof(void *),
+ GFP_NOFS);
+ if (!pointers) {
+ err = -ENOMEM;
+ goto cleanup_io;
+ }
+
+ faila = rbio->faila;
+ failb = rbio->failb;
+
+ if (rbio->read_rebuild) {
+ spin_lock_irq(&rbio->bio_list_lock);
+ set_bit(RBIO_RMW_LOCKED_BIT, &rbio->flags);
+ spin_unlock_irq(&rbio->bio_list_lock);
+ }
+
+ index_rbio_pages(rbio);
+
+ for (pagenr = 0; pagenr < nr_pages; pagenr++) {
+ /* setup our array of pointers with pages
+ * from each stripe
+ */
+ for (stripe = 0; stripe < rbio->bbio->num_stripes; stripe++) {
+ /*
+ * if we're rebuilding a read, we have to use
+ * pages from the bio list
+ */
+ if (rbio->read_rebuild &&
+ (stripe == faila || stripe == failb)) {
+ page = page_in_rbio(rbio, stripe, pagenr, 0);
+ } else {
+ page = rbio_stripe_page(rbio, stripe, pagenr);
+ }
+ pointers[stripe] = kmap(page);
+ }
+
+ /* all raid6 handling here */
+ if (rbio->raid_map[rbio->bbio->num_stripes - 1] ==
+ RAID6_Q_STRIPE) {
+
+ /*
+ * single failure, rebuild from parity raid5
+ * style
+ */
+ if (failb < 0) {
+ if (faila == rbio->nr_data) {
+ /*
+ * Just the P stripe has failed, without
+ * a bad data or Q stripe.
+ * TODO, we should redo the xor here.
+ */
+ err = -EIO;
+ goto cleanup;
+ }
+ /*
+ * a single failure in raid6 is rebuilt
+ * in the pstripe code below
+ */
+ goto pstripe;
+ }
+
+ /* make sure our ps and qs are in order */
+ if (faila > failb) {
+ int tmp = failb;
+ failb = faila;
+ faila = tmp;
+ }
+
+ /* if the q stripe is failed, do a pstripe reconstruction
+ * from the xors.
+ * If both the q stripe and the P stripe are failed, we're
+ * here due to a crc mismatch and we can't give them the
+ * data they want
+ */
+ if (rbio->raid_map[failb] == RAID6_Q_STRIPE) {
+ if (rbio->raid_map[faila] == RAID5_P_STRIPE) {
+ err = -EIO;
+ goto cleanup;
+ }
+ /*
+ * otherwise we have one bad data stripe and
+ * a good P stripe. raid5!
+ */
+ goto pstripe;
+ }
+
+ if (rbio->raid_map[failb] == RAID5_P_STRIPE) {
+ raid6_datap_recov(rbio->bbio->num_stripes,
+ PAGE_SIZE, faila, pointers);
+ } else {
+ raid6_2data_recov(rbio->bbio->num_stripes,
+ PAGE_SIZE, faila, failb,
+ pointers);
+ }
+ } else {
+ void *p;
+
+ /* rebuild from P stripe here (raid5 or raid6) */
+ BUG_ON(failb != -1);
+pstripe:
+ /* Copy parity block into failed block to start with */
+ memcpy(pointers[faila],
+ pointers[rbio->nr_data],
+ PAGE_CACHE_SIZE);
+
+ /* rearrange the pointer array */
+ p = pointers[faila];
+ for (stripe = faila; stripe < rbio->nr_data - 1; stripe++)
+ pointers[stripe] = pointers[stripe + 1];
+ pointers[rbio->nr_data - 1] = p;
+
+ /* xor in the rest */
+ run_xor(pointers, rbio->nr_data - 1, PAGE_CACHE_SIZE);
+ }
+ /* if we're doing this rebuild as part of an rmw, go through
+ * and set all of our private rbio pages in the
+ * failed stripes as uptodate. This way finish_rmw will
+ * know they can be trusted. If this was a read reconstruction,
+ * other endio functions will fiddle the uptodate bits
+ */
+ if (!rbio->read_rebuild) {
+ for (i = 0; i < nr_pages; i++) {
+ if (faila != -1) {
+ page = rbio_stripe_page(rbio, faila, i);
+ SetPageUptodate(page);
+ }
+ if (failb != -1) {
+ page = rbio_stripe_page(rbio, failb, i);
+ SetPageUptodate(page);
+ }
+ }
+ }
+ for (stripe = 0; stripe < rbio->bbio->num_stripes; stripe++) {
+ /*
+ * if we're rebuilding a read, we have to use
+ * pages from the bio list
+ */
+ if (rbio->read_rebuild &&
+ (stripe == faila || stripe == failb)) {
+ page = page_in_rbio(rbio, stripe, pagenr, 0);
+ } else {
+ page = rbio_stripe_page(rbio, stripe, pagenr);
+ }
+ kunmap(page);
+ }
+ }
+
+ err = 0;
+cleanup:
+ kfree(pointers);
+
+cleanup_io:
+
+ if (rbio->read_rebuild) {
+ if (err == 0)
+ cache_rbio_pages(rbio);
+ else
+ clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
+
+ rbio_orig_end_io(rbio, err, err == 0);
+ } else if (err == 0) {
+ rbio->faila = -1;
+ rbio->failb = -1;
+ finish_rmw(rbio);
+ } else {
+ rbio_orig_end_io(rbio, err, 0);
+ }
+}
+
+/*
+ * This is called only for stripes we've read from disk to
+ * reconstruct the parity.
+ */
+static void raid_recover_end_io(struct bio *bio, int err)
+{
+ struct btrfs_raid_bio *rbio = bio->bi_private;
+
+ /*
+ * we only read stripe pages off the disk, set them
+ * up to date if there were no errors
+ */
+ if (err)
+ fail_bio_stripe(rbio, bio);
+ else
+ set_bio_pages_uptodate(bio);
+ bio_put(bio);
+
+ if (!atomic_dec_and_test(&rbio->bbio->stripes_pending))
+ return;
+
+ if (atomic_read(&rbio->bbio->error) > rbio->bbio->max_errors)
+ rbio_orig_end_io(rbio, -EIO, 0);
+ else
+ __raid_recover_end_io(rbio);
+}
+
+/*
+ * reads everything we need off the disk to reconstruct
+ * the parity. endio handlers trigger final reconstruction
+ * when the IO is done.
+ *
+ * This is used both for reads from the higher layers and for
+ * parity construction required to finish a rmw cycle.
+ */
+static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
+{
+ int bios_to_read = 0;
+ struct btrfs_bio *bbio = rbio->bbio;
+ struct bio_list bio_list;
+ int ret;
+ int nr_pages = (rbio->stripe_len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+ int pagenr;
+ int stripe;
+ struct bio *bio;
+
+ bio_list_init(&bio_list);
+
+ ret = alloc_rbio_pages(rbio);
+ if (ret)
+ goto cleanup;
+
+ atomic_set(&rbio->bbio->error, 0);
+
+ /*
+ * read everything that hasn't failed. Thanks to the
+ * stripe cache, it is possible that some or all of these
+ * pages are going to be uptodate.
+ */
+ for (stripe = 0; stripe < bbio->num_stripes; stripe++) {
+ if (rbio->faila == stripe ||
+ rbio->failb == stripe)
+ continue;
+
+ for (pagenr = 0; pagenr < nr_pages; pagenr++) {
+ struct page *p;
+
+ /*
+ * the rmw code may have already read this
+ * page in
+ */
+ p = rbio_stripe_page(rbio, stripe, pagenr);
+ if (PageUptodate(p))
+ continue;
+
+ ret = rbio_add_io_page(rbio, &bio_list,
+ rbio_stripe_page(rbio, stripe, pagenr),
+ stripe, pagenr, rbio->stripe_len);
+ if (ret < 0)
+ goto cleanup;
+ }
+ }
+
+ bios_to_read = bio_list_size(&bio_list);
+ if (!bios_to_read) {
+ /*
+ * we might have no bios to read just because the pages
+ * were up to date, or we might have no bios to read because
+ * the devices were gone.
+ */
+ if (atomic_read(&rbio->bbio->error) <= rbio->bbio->max_errors) {
+ __raid_recover_end_io(rbio);
+ goto out;
+ } else {
+ goto cleanup;
+ }
+ }
+
+ /*
+ * the bbio may be freed once we submit the last bio. Make sure
+ * not to touch it after that
+ */
+ atomic_set(&bbio->stripes_pending, bios_to_read);
+ while (1) {
+ bio = bio_list_pop(&bio_list);
+ if (!bio)
+ break;
+
+ bio->bi_private = rbio;
+ bio->bi_end_io = raid_recover_end_io;
+
+ btrfs_bio_wq_end_io(rbio->fs_info, bio,
+ BTRFS_WQ_ENDIO_RAID56);
+
+ BUG_ON(!test_bit(BIO_UPTODATE, &bio->bi_flags));
+ submit_bio(READ, bio);
+ }
+out:
+ return 0;
+
+cleanup:
+ if (rbio->read_rebuild)
+ rbio_orig_end_io(rbio, -EIO, 0);
+ return -EIO;
+}
+
+/*
+ * the main entry point for reads from the higher layers. This
+ * is really only called when the normal read path had a failure,
+ * so we assume the bio they send down corresponds to a failed part
+ * of the drive.
+ */
+int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
+ struct btrfs_bio *bbio, u64 *raid_map,
+ u64 stripe_len, int mirror_num)
+{
+ struct btrfs_raid_bio *rbio;
+ int ret;
+
+ rbio = alloc_rbio(root, bbio, raid_map, stripe_len);
+ if (IS_ERR(rbio)) {
+ return PTR_ERR(rbio);
+ }
+
+ rbio->read_rebuild = 1;
+ bio_list_add(&rbio->bio_list, bio);
+ rbio->bio_list_bytes = bio->bi_size;
+
+ rbio->faila = find_logical_bio_stripe(rbio, bio);
+ if (rbio->faila == -1) {
+ BUG();
+ kfree(rbio);
+ return -EIO;
+ }
+
+ /*
+ * reconstruct from the q stripe if they are
+ * asking for mirror 3
+ */
+ if (mirror_num == 3)
+ rbio->failb = bbio->num_stripes - 2;
+
+ ret = lock_stripe_add(rbio);
+
+ /*
+ * __raid56_parity_recover will end the bio with
+ * any errors it hits. We don't want to return
+ * its error value up the stack because our caller
+ * will end up calling bio_endio with any nonzero
+ * return
+ */
+ if (ret == 0)
+ __raid56_parity_recover(rbio);
+ /*
+ * our rbio has been added to the list of
+ * rbios that will be handled after the
+ * currently lock owner is done
+ */
+ return 0;
+
+}
+
+static void rmw_work(struct btrfs_work *work)
+{
+ struct btrfs_raid_bio *rbio;
+
+ rbio = container_of(work, struct btrfs_raid_bio, work);
+ raid56_rmw_stripe(rbio);
+}
+
+static void read_rebuild_work(struct btrfs_work *work)
+{
+ struct btrfs_raid_bio *rbio;
+
+ rbio = container_of(work, struct btrfs_raid_bio, work);
+ __raid56_parity_recover(rbio);
+}
diff --git a/fs/btrfs/raid56.h b/fs/btrfs/raid56.h
new file mode 100644
index 00000000000..ea5d73bfdfb
--- /dev/null
+++ b/fs/btrfs/raid56.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2012 Fusion-io All rights reserved.
+ * Copyright (C) 2012 Intel Corp. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#ifndef __BTRFS_RAID56__
+#define __BTRFS_RAID56__
+static inline int nr_parity_stripes(struct map_lookup *map)
+{
+ if (map->type & BTRFS_BLOCK_GROUP_RAID5)
+ return 1;
+ else if (map->type & BTRFS_BLOCK_GROUP_RAID6)
+ return 2;
+ else
+ return 0;
+}
+
+static inline int nr_data_stripes(struct map_lookup *map)
+{
+ return map->num_stripes - nr_parity_stripes(map);
+}
+#define RAID5_P_STRIPE ((u64)-2)
+#define RAID6_Q_STRIPE ((u64)-1)
+
+#define is_parity_stripe(x) (((x) == RAID5_P_STRIPE) || \
+ ((x) == RAID6_Q_STRIPE))
+
+int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
+ struct btrfs_bio *bbio, u64 *raid_map,
+ u64 stripe_len, int mirror_num);
+int raid56_parity_write(struct btrfs_root *root, struct bio *bio,
+ struct btrfs_bio *bbio, u64 *raid_map,
+ u64 stripe_len);
+
+int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info);
+void btrfs_free_stripe_hash_table(struct btrfs_fs_info *info);
+#endif
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 96b93daa0bb..1031b69252c 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -955,10 +955,11 @@ int btrfs_reada_wait(void *handle)
while (atomic_read(&rc->elems)) {
wait_event_timeout(rc->wait, atomic_read(&rc->elems) == 0,
5 * HZ);
- dump_devs(rc->root->fs_info, rc->elems < 10 ? 1 : 0);
+ dump_devs(rc->root->fs_info,
+ atomic_read(&rc->elems) < 10 ? 1 : 0);
}
- dump_devs(rc->root->fs_info, rc->elems < 10 ? 1 : 0);
+ dump_devs(rc->root->fs_info, atomic_read(&rc->elems) < 10 ? 1 : 0);
kref_put(&rc->refcnt, reada_control_release);
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 300e09ac365..704a1b8d2a2 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -326,8 +326,7 @@ static struct rb_node *tree_search(struct rb_root *root, u64 bytenr)
return NULL;
}
-void backref_tree_panic(struct rb_node *rb_node, int errno,
- u64 bytenr)
+static void backref_tree_panic(struct rb_node *rb_node, int errno, u64 bytenr)
{
struct btrfs_fs_info *fs_info = NULL;
@@ -619,10 +618,13 @@ static noinline_for_stack
int find_inline_backref(struct extent_buffer *leaf, int slot,
unsigned long *ptr, unsigned long *end)
{
+ struct btrfs_key key;
struct btrfs_extent_item *ei;
struct btrfs_tree_block_info *bi;
u32 item_size;
+ btrfs_item_key_to_cpu(leaf, &key, slot);
+
item_size = btrfs_item_size_nr(leaf, slot);
#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
if (item_size < sizeof(*ei)) {
@@ -634,13 +636,18 @@ int find_inline_backref(struct extent_buffer *leaf, int slot,
WARN_ON(!(btrfs_extent_flags(leaf, ei) &
BTRFS_EXTENT_FLAG_TREE_BLOCK));
- if (item_size <= sizeof(*ei) + sizeof(*bi)) {
+ if (key.type == BTRFS_EXTENT_ITEM_KEY &&
+ item_size <= sizeof(*ei) + sizeof(*bi)) {
WARN_ON(item_size < sizeof(*ei) + sizeof(*bi));
return 1;
}
- bi = (struct btrfs_tree_block_info *)(ei + 1);
- *ptr = (unsigned long)(bi + 1);
+ if (key.type == BTRFS_EXTENT_ITEM_KEY) {
+ bi = (struct btrfs_tree_block_info *)(ei + 1);
+ *ptr = (unsigned long)(bi + 1);
+ } else {
+ *ptr = (unsigned long)(ei + 1);
+ }
*end = (unsigned long)ei + item_size;
return 0;
}
@@ -708,7 +715,7 @@ again:
end = 0;
ptr = 0;
key.objectid = cur->bytenr;
- key.type = BTRFS_EXTENT_ITEM_KEY;
+ key.type = BTRFS_METADATA_ITEM_KEY;
key.offset = (u64)-1;
path1->search_commit_root = 1;
@@ -766,7 +773,8 @@ again:
break;
}
- if (key.type == BTRFS_EXTENT_ITEM_KEY) {
+ if (key.type == BTRFS_EXTENT_ITEM_KEY ||
+ key.type == BTRFS_METADATA_ITEM_KEY) {
ret = find_inline_backref(eb, path1->slots[0],
&ptr, &end);
if (ret)
@@ -1269,6 +1277,8 @@ static int __update_reloc_root(struct btrfs_root *root, int del)
}
spin_unlock(&rc->reloc_root_tree.lock);
+ if (!node)
+ return 0;
BUG_ON((struct btrfs_root *)node->data != root);
if (!del) {
@@ -1760,7 +1770,11 @@ again:
eb = read_tree_block(dest, old_bytenr, blocksize,
old_ptr_gen);
- BUG_ON(!eb);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ ret = (!eb) ? -ENOMEM : -EIO;
+ free_extent_buffer(eb);
+ return ret;
+ }
btrfs_tree_lock(eb);
if (cow) {
ret = btrfs_cow_block(trans, dest, eb, parent,
@@ -1913,6 +1927,10 @@ int walk_down_reloc_tree(struct btrfs_root *root, struct btrfs_path *path,
bytenr = btrfs_node_blockptr(eb, path->slots[i]);
blocksize = btrfs_level_size(root, i - 1);
eb = read_tree_block(root, bytenr, blocksize, ptr_gen);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
+ return -EIO;
+ }
BUG_ON(btrfs_header_level(eb) != i - 1);
path->nodes[i - 1] = eb;
path->slots[i - 1] = 0;
@@ -2238,13 +2256,28 @@ again:
}
static noinline_for_stack
+void free_reloc_roots(struct list_head *list)
+{
+ struct btrfs_root *reloc_root;
+
+ while (!list_empty(list)) {
+ reloc_root = list_entry(list->next, struct btrfs_root,
+ root_list);
+ __update_reloc_root(reloc_root, 1);
+ free_extent_buffer(reloc_root->node);
+ free_extent_buffer(reloc_root->commit_root);
+ kfree(reloc_root);
+ }
+}
+
+static noinline_for_stack
int merge_reloc_roots(struct reloc_control *rc)
{
struct btrfs_root *root;
struct btrfs_root *reloc_root;
LIST_HEAD(reloc_roots);
int found = 0;
- int ret;
+ int ret = 0;
again:
root = rc->extent_root;
@@ -2270,20 +2303,33 @@ again:
BUG_ON(root->reloc_root != reloc_root);
ret = merge_reloc_root(rc, root);
- BUG_ON(ret);
+ if (ret)
+ goto out;
} else {
list_del_init(&reloc_root->root_list);
}
ret = btrfs_drop_snapshot(reloc_root, rc->block_rsv, 0, 1);
- BUG_ON(ret < 0);
+ if (ret < 0) {
+ if (list_empty(&reloc_root->root_list))
+ list_add_tail(&reloc_root->root_list,
+ &reloc_roots);
+ goto out;
+ }
}
if (found) {
found = 0;
goto again;
}
+out:
+ if (ret) {
+ btrfs_std_error(root->fs_info, ret);
+ if (!list_empty(&reloc_roots))
+ free_reloc_roots(&reloc_roots);
+ }
+
BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root));
- return 0;
+ return ret;
}
static void free_block_list(struct rb_root *blocks)
@@ -2562,7 +2608,8 @@ static int do_relocation(struct btrfs_trans_handle *trans,
blocksize = btrfs_level_size(root, node->level);
generation = btrfs_node_ptr_generation(upper->eb, slot);
eb = read_tree_block(root, bytenr, blocksize, generation);
- if (!eb) {
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
err = -EIO;
goto next;
}
@@ -2723,7 +2770,10 @@ static int get_tree_block_key(struct reloc_control *rc,
BUG_ON(block->key_ready);
eb = read_tree_block(rc->extent_root, block->bytenr,
block->key.objectid, block->key.offset);
- BUG_ON(!eb);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ free_extent_buffer(eb);
+ return -EIO;
+ }
WARN_ON(btrfs_header_level(eb) != block->level);
if (block->level == 0)
btrfs_item_key_to_cpu(eb, &block->key, 0);
@@ -2738,8 +2788,13 @@ static int reada_tree_block(struct reloc_control *rc,
struct tree_block *block)
{
BUG_ON(block->key_ready);
- readahead_tree_block(rc->extent_root, block->bytenr,
- block->key.objectid, block->key.offset);
+ if (block->key.type == BTRFS_METADATA_ITEM_KEY)
+ readahead_tree_block(rc->extent_root, block->bytenr,
+ block->key.objectid,
+ rc->extent_root->leafsize);
+ else
+ readahead_tree_block(rc->extent_root, block->bytenr,
+ block->key.objectid, block->key.offset);
return 0;
}
@@ -2818,8 +2873,10 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
int err = 0;
path = btrfs_alloc_path();
- if (!path)
- return -ENOMEM;
+ if (!path) {
+ err = -ENOMEM;
+ goto out_free_blocks;
+ }
rb_node = rb_first(blocks);
while (rb_node) {
@@ -2832,8 +2889,11 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
rb_node = rb_first(blocks);
while (rb_node) {
block = rb_entry(rb_node, struct tree_block, rb_node);
- if (!block->key_ready)
- get_tree_block_key(rc, block);
+ if (!block->key_ready) {
+ err = get_tree_block_key(rc, block);
+ if (err)
+ goto out_free_path;
+ }
rb_node = rb_next(rb_node);
}
@@ -2858,10 +2918,12 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
rb_node = rb_next(rb_node);
}
out:
- free_block_list(blocks);
err = finish_pending_nodes(trans, rc, path, err);
+out_free_path:
btrfs_free_path(path);
+out_free_blocks:
+ free_block_list(blocks);
return err;
}
@@ -2932,7 +2994,7 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end,
lock_extent(&BTRFS_I(inode)->io_tree, start, end);
while (1) {
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
+ ret = add_extent_mapping(em_tree, em, 0);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
@@ -3017,7 +3079,7 @@ static int relocate_file_extent_cluster(struct inode *inode,
}
}
- page_start = (u64)page->index << PAGE_CACHE_SHIFT;
+ page_start = page_offset(page);
page_end = page_start + PAGE_CACHE_SIZE - 1;
lock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end);
@@ -3143,12 +3205,17 @@ static int add_tree_block(struct reloc_control *rc,
eb = path->nodes[0];
item_size = btrfs_item_size_nr(eb, path->slots[0]);
- if (item_size >= sizeof(*ei) + sizeof(*bi)) {
+ if (extent_key->type == BTRFS_METADATA_ITEM_KEY ||
+ item_size >= sizeof(*ei) + sizeof(*bi)) {
ei = btrfs_item_ptr(eb, path->slots[0],
struct btrfs_extent_item);
- bi = (struct btrfs_tree_block_info *)(ei + 1);
+ if (extent_key->type == BTRFS_EXTENT_ITEM_KEY) {
+ bi = (struct btrfs_tree_block_info *)(ei + 1);
+ level = btrfs_tree_block_level(eb, bi);
+ } else {
+ level = (int)extent_key->offset;
+ }
generation = btrfs_extent_generation(eb, ei);
- level = btrfs_tree_block_level(eb, bi);
} else {
#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
u64 ref_owner;
@@ -3177,7 +3244,7 @@ static int add_tree_block(struct reloc_control *rc,
return -ENOMEM;
block->bytenr = extent_key->objectid;
- block->key.objectid = extent_key->offset;
+ block->key.objectid = rc->extent_root->leafsize;
block->key.offset = generation;
block->level = level;
block->key_ready = 0;
@@ -3219,9 +3286,15 @@ static int __add_tree_block(struct reloc_control *rc,
ret = btrfs_search_slot(NULL, rc->extent_root, &key, path, 0, 0);
if (ret < 0)
goto out;
- BUG_ON(ret);
btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+ if (ret > 0) {
+ if (key.objectid == bytenr &&
+ key.type == BTRFS_METADATA_ITEM_KEY)
+ ret = 0;
+ }
+ BUG_ON(ret);
+
ret = add_tree_block(rc, &key, path, blocks);
out:
btrfs_free_path(path);
@@ -3242,7 +3315,8 @@ static int block_use_full_backref(struct reloc_control *rc,
return 1;
ret = btrfs_lookup_extent_info(NULL, rc->extent_root,
- eb->start, eb->len, NULL, &flags);
+ eb->start, btrfs_header_level(eb), 1,
+ NULL, &flags);
BUG_ON(ret);
if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)
@@ -3472,7 +3546,7 @@ out:
}
/*
- * hepler to find all tree blocks that reference a given data extent
+ * helper to find all tree blocks that reference a given data extent
*/
static noinline_for_stack
int add_data_references(struct reloc_control *rc,
@@ -3566,7 +3640,7 @@ int add_data_references(struct reloc_control *rc,
}
/*
- * hepler to find next unprocessed extent
+ * helper to find next unprocessed extent
*/
static noinline_for_stack
int find_next_extent(struct btrfs_trans_handle *trans,
@@ -3611,12 +3685,25 @@ next:
break;
}
- if (key.type != BTRFS_EXTENT_ITEM_KEY ||
+ if (key.type != BTRFS_EXTENT_ITEM_KEY &&
+ key.type != BTRFS_METADATA_ITEM_KEY) {
+ path->slots[0]++;
+ goto next;
+ }
+
+ if (key.type == BTRFS_EXTENT_ITEM_KEY &&
key.objectid + key.offset <= rc->search_start) {
path->slots[0]++;
goto next;
}
+ if (key.type == BTRFS_METADATA_ITEM_KEY &&
+ key.objectid + rc->extent_root->leafsize <=
+ rc->search_start) {
+ path->slots[0]++;
+ goto next;
+ }
+
ret = find_first_extent_bit(&rc->processed_blocks,
key.objectid, &start, &end,
EXTENT_DIRTY, NULL);
@@ -3625,7 +3712,11 @@ next:
btrfs_release_path(path);
rc->search_start = end + 1;
} else {
- rc->search_start = key.objectid + key.offset;
+ if (key.type == BTRFS_EXTENT_ITEM_KEY)
+ rc->search_start = key.objectid + key.offset;
+ else
+ rc->search_start = key.objectid +
+ rc->extent_root->leafsize;
memcpy(extent_key, &key, sizeof(key));
return 0;
}
@@ -3698,7 +3789,15 @@ int prepare_to_relocate(struct reloc_control *rc)
set_reloc_control(rc);
trans = btrfs_join_transaction(rc->extent_root);
- BUG_ON(IS_ERR(trans));
+ if (IS_ERR(trans)) {
+ unset_reloc_control(rc);
+ /*
+ * extent tree is not a ref_cow tree and has no reloc_root to
+ * cleanup. And callers are responsible to free the above
+ * block rsv.
+ */
+ return PTR_ERR(trans);
+ }
btrfs_commit_transaction(trans, rc->extent_root);
return 0;
}
@@ -3730,7 +3829,11 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
while (1) {
progress++;
trans = btrfs_start_transaction(rc->extent_root, 0);
- BUG_ON(IS_ERR(trans));
+ if (IS_ERR(trans)) {
+ err = PTR_ERR(trans);
+ trans = NULL;
+ break;
+ }
restart:
if (update_backref_cache(trans, &rc->backref_cache)) {
btrfs_end_transaction(trans, rc->extent_root);
@@ -4060,10 +4163,7 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
while (1) {
mutex_lock(&fs_info->cleaner_mutex);
-
- btrfs_clean_old_snapshots(fs_info->tree_root);
ret = relocate_block_group(rc);
-
mutex_unlock(&fs_info->cleaner_mutex);
if (ret < 0) {
err = ret;
@@ -4264,14 +4364,9 @@ int btrfs_recover_relocation(struct btrfs_root *root)
out_free:
kfree(rc);
out:
- while (!list_empty(&reloc_roots)) {
- reloc_root = list_entry(reloc_roots.next,
- struct btrfs_root, root_list);
- list_del(&reloc_root->root_list);
- free_extent_buffer(reloc_root->node);
- free_extent_buffer(reloc_root->commit_root);
- kfree(reloc_root);
- }
+ if (!list_empty(&reloc_roots))
+ free_reloc_roots(&reloc_roots);
+
btrfs_free_path(path);
if (err == 0) {
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 668af537a3e..5bf1ed57f17 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -29,9 +29,8 @@
* generation numbers as then we know the root was once mounted with an older
* kernel that was not aware of the root item structure change.
*/
-void btrfs_read_root_item(struct btrfs_root *root,
- struct extent_buffer *eb, int slot,
- struct btrfs_root_item *item)
+void btrfs_read_root_item(struct extent_buffer *eb, int slot,
+ struct btrfs_root_item *item)
{
uuid_le uuid;
int len;
@@ -104,7 +103,7 @@ int btrfs_find_last_root(struct btrfs_root *root, u64 objectid,
goto out;
}
if (item)
- btrfs_read_root_item(root, l, slot, item);
+ btrfs_read_root_item(l, slot, item);
if (key)
memcpy(key, &found_key, sizeof(found_key));
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 67783e03d12..f489e24659a 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -28,6 +28,7 @@
#include "dev-replace.h"
#include "check-integrity.h"
#include "rcu-string.h"
+#include "raid56.h"
/*
* This is only the first step towards a full-features scrub. It reads all
@@ -541,7 +542,6 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
eb = path->nodes[0];
ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
item_size = btrfs_item_size_nr(eb, path->slots[0]);
- btrfs_release_path(path);
if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
do {
@@ -557,7 +557,9 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
ret < 0 ? -1 : ref_level,
ret < 0 ? -1 : ref_root);
} while (ret != 1);
+ btrfs_release_path(path);
} else {
+ btrfs_release_path(path);
swarn.path = path;
swarn.dev = dev;
iterate_extent_inodes(fs_info, found_key.objectid,
@@ -1334,7 +1336,6 @@ static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
int page_num;
u8 calculated_csum[BTRFS_CSUM_SIZE];
u32 crc = ~(u32)0;
- struct btrfs_root *root = fs_info->extent_root;
void *mapped_buffer;
WARN_ON(!sblock->pagev[0]->page);
@@ -1363,12 +1364,11 @@ static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
for (page_num = 0;;) {
if (page_num == 0 && is_metadata)
- crc = btrfs_csum_data(root,
+ crc = btrfs_csum_data(
((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE,
crc, PAGE_SIZE - BTRFS_CSUM_SIZE);
else
- crc = btrfs_csum_data(root, mapped_buffer, crc,
- PAGE_SIZE);
+ crc = btrfs_csum_data(mapped_buffer, crc, PAGE_SIZE);
kunmap_atomic(mapped_buffer);
page_num++;
@@ -1655,7 +1655,6 @@ static int scrub_checksum_data(struct scrub_block *sblock)
void *buffer;
u32 crc = ~(u32)0;
int fail = 0;
- struct btrfs_root *root = sctx->dev_root;
u64 len;
int index;
@@ -1672,7 +1671,7 @@ static int scrub_checksum_data(struct scrub_block *sblock)
for (;;) {
u64 l = min_t(u64, len, PAGE_SIZE);
- crc = btrfs_csum_data(root, buffer, crc, l);
+ crc = btrfs_csum_data(buffer, crc, l);
kunmap_atomic(buffer);
len -= l;
if (len == 0)
@@ -1742,7 +1741,7 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock)
for (;;) {
u64 l = min_t(u64, len, mapped_size);
- crc = btrfs_csum_data(root, p, crc, l);
+ crc = btrfs_csum_data(p, crc, l);
kunmap_atomic(mapped_buffer);
len -= l;
if (len == 0)
@@ -1803,7 +1802,7 @@ static int scrub_checksum_super(struct scrub_block *sblock)
for (;;) {
u64 l = min_t(u64, len, mapped_size);
- crc = btrfs_csum_data(root, p, crc, l);
+ crc = btrfs_csum_data(p, crc, l);
kunmap_atomic(mapped_buffer);
len -= l;
if (len == 0)
@@ -2234,12 +2233,12 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
u64 flags;
int ret;
int slot;
- int i;
u64 nstripes;
struct extent_buffer *l;
struct btrfs_key key;
u64 physical;
u64 logical;
+ u64 logic_end;
u64 generation;
int mirror_num;
struct reada_control *reada1;
@@ -2253,6 +2252,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
u64 extent_len;
struct btrfs_device *extent_dev;
int extent_mirror_num;
+ int stop_loop;
+
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ if (num >= nr_data_stripes(map)) {
+ return 0;
+ }
+ }
nstripes = length;
offset = 0;
@@ -2306,8 +2313,8 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
key_start.type = BTRFS_EXTENT_ITEM_KEY;
key_start.offset = (u64)0;
key_end.objectid = base + offset + nstripes * increment;
- key_end.type = BTRFS_EXTENT_ITEM_KEY;
- key_end.offset = (u64)0;
+ key_end.type = BTRFS_METADATA_ITEM_KEY;
+ key_end.offset = (u64)-1;
reada1 = btrfs_reada_add(root, &key_start, &key_end);
key_start.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
@@ -2345,8 +2352,9 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
*/
logical = base + offset;
physical = map->stripes[num].physical;
+ logic_end = logical + increment * nstripes;
ret = 0;
- for (i = 0; i < nstripes; ++i) {
+ while (logical < logic_end) {
/*
* canceled?
*/
@@ -2382,19 +2390,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
wake_up(&fs_info->scrub_pause_wait);
}
- ret = btrfs_lookup_csums_range(csum_root, logical,
- logical + map->stripe_len - 1,
- &sctx->csum_list, 1);
- if (ret)
- goto out;
-
key.objectid = logical;
key.type = BTRFS_EXTENT_ITEM_KEY;
- key.offset = (u64)0;
+ key.offset = (u64)-1;
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
goto out;
+
if (ret > 0) {
ret = btrfs_previous_item(root, path, 0,
BTRFS_EXTENT_ITEM_KEY);
@@ -2411,7 +2414,10 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
}
}
+ stop_loop = 0;
while (1) {
+ u64 bytes;
+
l = path->nodes[0];
slot = path->slots[0];
if (slot >= btrfs_header_nritems(l)) {
@@ -2421,19 +2427,30 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
if (ret < 0)
goto out;
+ stop_loop = 1;
break;
}
btrfs_item_key_to_cpu(l, &key, slot);
- if (key.objectid + key.offset <= logical)
- goto next;
+ if (key.type == BTRFS_METADATA_ITEM_KEY)
+ bytes = root->leafsize;
+ else
+ bytes = key.offset;
- if (key.objectid >= logical + map->stripe_len)
- break;
+ if (key.objectid + bytes <= logical)
+ goto next;
- if (btrfs_key_type(&key) != BTRFS_EXTENT_ITEM_KEY)
+ if (key.type != BTRFS_EXTENT_ITEM_KEY &&
+ key.type != BTRFS_METADATA_ITEM_KEY)
goto next;
+ if (key.objectid >= logical + map->stripe_len) {
+ /* out of this device extent */
+ if (key.objectid >= logic_end)
+ stop_loop = 1;
+ break;
+ }
+
extent = btrfs_item_ptr(l, slot,
struct btrfs_extent_item);
flags = btrfs_extent_flags(l, extent);
@@ -2449,22 +2466,24 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
goto next;
}
+again:
+ extent_logical = key.objectid;
+ extent_len = bytes;
+
/*
* trim extent to this stripe
*/
- if (key.objectid < logical) {
- key.offset -= logical - key.objectid;
- key.objectid = logical;
+ if (extent_logical < logical) {
+ extent_len -= logical - extent_logical;
+ extent_logical = logical;
}
- if (key.objectid + key.offset >
+ if (extent_logical + extent_len >
logical + map->stripe_len) {
- key.offset = logical + map->stripe_len -
- key.objectid;
+ extent_len = logical + map->stripe_len -
+ extent_logical;
}
- extent_logical = key.objectid;
- extent_physical = key.objectid - logical + physical;
- extent_len = key.offset;
+ extent_physical = extent_logical - logical + physical;
extent_dev = scrub_dev;
extent_mirror_num = mirror_num;
if (is_dev_replace)
@@ -2472,13 +2491,35 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
extent_len, &extent_physical,
&extent_dev,
&extent_mirror_num);
+
+ ret = btrfs_lookup_csums_range(csum_root, logical,
+ logical + map->stripe_len - 1,
+ &sctx->csum_list, 1);
+ if (ret)
+ goto out;
+
ret = scrub_extent(sctx, extent_logical, extent_len,
extent_physical, extent_dev, flags,
generation, extent_mirror_num,
- key.objectid - logical + physical);
+ extent_physical);
if (ret)
goto out;
+ if (extent_logical + extent_len <
+ key.objectid + bytes) {
+ logical += increment;
+ physical += map->stripe_len;
+
+ if (logical < key.objectid + bytes) {
+ cond_resched();
+ goto again;
+ }
+
+ if (logical >= logic_end) {
+ stop_loop = 1;
+ break;
+ }
+ }
next:
path->slots[0]++;
}
@@ -2486,8 +2527,14 @@ next:
logical += increment;
physical += map->stripe_len;
spin_lock(&sctx->stat_lock);
- sctx->stat.last_physical = physical;
+ if (stop_loop)
+ sctx->stat.last_physical = map->stripes[num].physical +
+ length;
+ else
+ sctx->stat.last_physical = physical;
spin_unlock(&sctx->stat_lock);
+ if (stop_loop)
+ break;
}
out:
/* push queued extents */
@@ -2708,7 +2755,7 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx,
int ret;
struct btrfs_root *root = sctx->dev_root;
- if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
+ if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
return -EIO;
gen = root->fs_info->last_trans_committed;
@@ -2996,28 +3043,6 @@ int btrfs_scrub_cancel_dev(struct btrfs_fs_info *fs_info,
return 0;
}
-int btrfs_scrub_cancel_devid(struct btrfs_root *root, u64 devid)
-{
- struct btrfs_fs_info *fs_info = root->fs_info;
- struct btrfs_device *dev;
- int ret;
-
- /*
- * we have to hold the device_list_mutex here so the device
- * does not go away in cancel_dev. FIXME: find a better solution
- */
- mutex_lock(&fs_info->fs_devices->device_list_mutex);
- dev = btrfs_find_device(fs_info, devid, NULL, NULL);
- if (!dev) {
- mutex_unlock(&fs_info->fs_devices->device_list_mutex);
- return -ENODEV;
- }
- ret = btrfs_scrub_cancel_dev(fs_info, dev);
- mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-
- return ret;
-}
-
int btrfs_scrub_progress(struct btrfs_root *root, u64 devid,
struct btrfs_scrub_progress *progress)
{
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 321b7fb4e44..ff40f1c00ce 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -85,6 +85,7 @@ struct send_ctx {
u32 send_max_size;
u64 total_send_size;
u64 cmd_send_size[BTRFS_SEND_C_MAX + 1];
+ u64 flags; /* 'flags' member of btrfs_ioctl_send_args is u64 */
struct vfsmount *mnt;
@@ -386,7 +387,7 @@ static struct btrfs_path *alloc_path_for_send(void)
return path;
}
-int write_buf(struct file *filp, const void *buf, u32 len, loff_t *off)
+static int write_buf(struct file *filp, const void *buf, u32 len, loff_t *off)
{
int ret;
mm_segment_t old_fs;
@@ -3478,7 +3479,6 @@ static int __process_changed_new_xattr(int num, struct btrfs_key *di_key,
struct send_ctx *sctx = ctx;
char *found_data = NULL;
int found_data_len = 0;
- struct fs_path *p = NULL;
ret = find_xattr(sctx, sctx->parent_root, sctx->right_path,
sctx->cmp_key, name, name_len, &found_data,
@@ -3497,7 +3497,6 @@ static int __process_changed_new_xattr(int num, struct btrfs_key *di_key,
}
kfree(found_data);
- fs_path_free(sctx, p);
return ret;
}
@@ -3709,6 +3708,39 @@ out:
return ret;
}
+/*
+ * Send an update extent command to user space.
+ */
+static int send_update_extent(struct send_ctx *sctx,
+ u64 offset, u32 len)
+{
+ int ret = 0;
+ struct fs_path *p;
+
+ p = fs_path_alloc(sctx);
+ if (!p)
+ return -ENOMEM;
+
+ ret = begin_cmd(sctx, BTRFS_SEND_C_UPDATE_EXTENT);
+ if (ret < 0)
+ goto out;
+
+ ret = get_cur_path(sctx, sctx->cur_ino, sctx->cur_inode_gen, p);
+ if (ret < 0)
+ goto out;
+
+ TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p);
+ TLV_PUT_U64(sctx, BTRFS_SEND_A_FILE_OFFSET, offset);
+ TLV_PUT_U64(sctx, BTRFS_SEND_A_SIZE, len);
+
+ ret = send_cmd(sctx);
+
+tlv_put_failure:
+out:
+ fs_path_free(sctx, p);
+ return ret;
+}
+
static int send_write_or_clone(struct send_ctx *sctx,
struct btrfs_path *path,
struct btrfs_key *key,
@@ -3744,7 +3776,11 @@ static int send_write_or_clone(struct send_ctx *sctx,
goto out;
}
- if (!clone_root) {
+ if (clone_root) {
+ ret = send_clone(sctx, offset, len, clone_root);
+ } else if (sctx->flags & BTRFS_SEND_FLAG_NO_FILE_DATA) {
+ ret = send_update_extent(sctx, offset, len);
+ } else {
while (pos < len) {
l = len - pos;
if (l > BTRFS_SEND_READ_SIZE)
@@ -3757,10 +3793,7 @@ static int send_write_or_clone(struct send_ctx *sctx,
pos += ret;
}
ret = 0;
- } else {
- ret = send_clone(sctx, offset, len, clone_root);
}
-
out:
return ret;
}
@@ -3910,12 +3943,10 @@ static int is_extent_unchanged(struct send_ctx *sctx,
found_key.type != key.type) {
key.offset += right_len;
break;
- } else {
- if (found_key.offset != key.offset + right_len) {
- /* Should really not happen */
- ret = -EIO;
- goto out;
- }
+ }
+ if (found_key.offset != key.offset + right_len) {
+ ret = 0;
+ goto out;
}
key = found_key;
}
@@ -4496,9 +4527,11 @@ static int send_subvol(struct send_ctx *sctx)
{
int ret;
- ret = send_header(sctx);
- if (ret < 0)
- goto out;
+ if (!(sctx->flags & BTRFS_SEND_FLAG_OMIT_STREAM_HEADER)) {
+ ret = send_header(sctx);
+ if (ret < 0)
+ goto out;
+ }
ret = send_subvol_begin(sctx);
if (ret < 0)
@@ -4536,7 +4569,6 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
struct btrfs_fs_info *fs_info;
struct btrfs_ioctl_send_args *arg = NULL;
struct btrfs_key key;
- struct file *filp = NULL;
struct send_ctx *sctx = NULL;
u32 i;
u64 *clone_sources_tmp = NULL;
@@ -4544,7 +4576,7 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- send_root = BTRFS_I(fdentry(mnt_file)->d_inode)->root;
+ send_root = BTRFS_I(file_inode(mnt_file))->root;
fs_info = send_root->fs_info;
arg = memdup_user(arg_, sizeof(*arg));
@@ -4561,6 +4593,11 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
goto out;
}
+ if (arg->flags & ~BTRFS_SEND_FLAG_MASK) {
+ ret = -EINVAL;
+ goto out;
+ }
+
sctx = kzalloc(sizeof(struct send_ctx), GFP_NOFS);
if (!sctx) {
ret = -ENOMEM;
@@ -4572,9 +4609,11 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
INIT_RADIX_TREE(&sctx->name_cache, GFP_NOFS);
INIT_LIST_HEAD(&sctx->name_cache_list);
+ sctx->flags = arg->flags;
+
sctx->send_filp = fget(arg->send_fd);
- if (IS_ERR(sctx->send_filp)) {
- ret = PTR_ERR(sctx->send_filp);
+ if (!sctx->send_filp) {
+ ret = -EBADF;
goto out;
}
@@ -4665,16 +4704,16 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
if (ret < 0)
goto out;
- ret = begin_cmd(sctx, BTRFS_SEND_C_END);
- if (ret < 0)
- goto out;
- ret = send_cmd(sctx);
- if (ret < 0)
- goto out;
+ if (!(sctx->flags & BTRFS_SEND_FLAG_OMIT_END_CMD)) {
+ ret = begin_cmd(sctx, BTRFS_SEND_C_END);
+ if (ret < 0)
+ goto out;
+ ret = send_cmd(sctx);
+ if (ret < 0)
+ goto out;
+ }
out:
- if (filp)
- fput(filp);
kfree(arg);
vfree(clone_sources_tmp);
diff --git a/fs/btrfs/send.h b/fs/btrfs/send.h
index 1bf4f32fd4e..48d425aef05 100644
--- a/fs/btrfs/send.h
+++ b/fs/btrfs/send.h
@@ -86,6 +86,7 @@ enum btrfs_send_cmd {
BTRFS_SEND_C_UTIMES,
BTRFS_SEND_C_END,
+ BTRFS_SEND_C_UPDATE_EXTENT,
__BTRFS_SEND_C_MAX,
};
#define BTRFS_SEND_C_MAX (__BTRFS_SEND_C_MAX - 1)
@@ -130,5 +131,4 @@ enum {
#ifdef __KERNEL__
long btrfs_ioctl_send(struct file *mnt_file, void __user *arg);
-int write_buf(struct file *filp, const void *buf, u32 len, loff_t *off);
#endif
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index d8982e9601d..a4807ced23c 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -41,13 +41,13 @@
#include <linux/slab.h>
#include <linux/cleancache.h>
#include <linux/ratelimit.h>
+#include <linux/btrfs.h>
#include "compat.h"
#include "delayed-inode.h"
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
-#include "ioctl.h"
#include "print-tree.h"
#include "xattr.h"
#include "volumes.h"
@@ -56,6 +56,7 @@
#include "compression.h"
#include "rcu-string.h"
#include "dev-replace.h"
+#include "free-space-cache.h"
#define CREATE_TRACE_POINTS
#include <trace/events/btrfs.h>
@@ -63,10 +64,9 @@
static const struct super_operations btrfs_super_ops;
static struct file_system_type btrfs_fs_type;
-static const char *btrfs_decode_error(struct btrfs_fs_info *fs_info, int errno,
- char nbuf[16])
+static const char *btrfs_decode_error(int errno)
{
- char *errstr = NULL;
+ char *errstr = "unknown";
switch (errno) {
case -EIO:
@@ -81,29 +81,24 @@ static const char *btrfs_decode_error(struct btrfs_fs_info *fs_info, int errno,
case -EEXIST:
errstr = "Object already exists";
break;
- default:
- if (nbuf) {
- if (snprintf(nbuf, 16, "error %d", -errno) >= 0)
- errstr = nbuf;
- }
+ case -ENOSPC:
+ errstr = "No space left";
+ break;
+ case -ENOENT:
+ errstr = "No such entry";
break;
}
return errstr;
}
-static void __save_error_info(struct btrfs_fs_info *fs_info)
+static void save_error_info(struct btrfs_fs_info *fs_info)
{
/*
* today we only save the error info into ram. Long term we'll
* also send it down to the disk
*/
- fs_info->fs_state = BTRFS_SUPER_FLAG_ERROR;
-}
-
-static void save_error_info(struct btrfs_fs_info *fs_info)
-{
- __save_error_info(fs_info);
+ set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
}
/* btrfs handle error by forcing the filesystem readonly */
@@ -114,9 +109,9 @@ static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
if (sb->s_flags & MS_RDONLY)
return;
- if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+ if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
sb->s_flags |= MS_RDONLY;
- printk(KERN_INFO "btrfs is forced readonly\n");
+ btrfs_info(fs_info, "forced readonly");
/*
* Note that a running device replace operation is not
* canceled here although there is no way to update
@@ -127,7 +122,6 @@ static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
* mounted writeable again, the device replace
* operation continues.
*/
-// WARN_ON(1);
}
}
@@ -140,10 +134,7 @@ void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
unsigned int line, int errno, const char *fmt, ...)
{
struct super_block *sb = fs_info->sb;
- char nbuf[16];
const char *errstr;
- va_list args;
- va_start(args, fmt);
/*
* Special case: if the error is EROFS, and we're already
@@ -152,26 +143,27 @@ void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
if (errno == -EROFS && (sb->s_flags & MS_RDONLY))
return;
- errstr = btrfs_decode_error(fs_info, errno, nbuf);
+ errstr = btrfs_decode_error(errno);
if (fmt) {
- struct va_format vaf = {
- .fmt = fmt,
- .va = &args,
- };
+ struct va_format vaf;
+ va_list args;
- printk(KERN_CRIT "BTRFS error (device %s) in %s:%d: %s (%pV)\n",
- sb->s_id, function, line, errstr, &vaf);
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ printk(KERN_CRIT "BTRFS error (device %s) in %s:%d: errno=%d %s (%pV)\n",
+ sb->s_id, function, line, errno, errstr, &vaf);
+ va_end(args);
} else {
- printk(KERN_CRIT "BTRFS error (device %s) in %s:%d: %s\n",
- sb->s_id, function, line, errstr);
+ printk(KERN_CRIT "BTRFS error (device %s) in %s:%d: errno=%d %s\n",
+ sb->s_id, function, line, errno, errstr);
}
/* Don't go through full error handling during mount */
- if (sb->s_flags & MS_BORN) {
- save_error_info(fs_info);
+ save_error_info(fs_info);
+ if (sb->s_flags & MS_BORN)
btrfs_handle_error(fs_info);
- }
- va_end(args);
}
static const char * const logtypes[] = {
@@ -185,7 +177,7 @@ static const char * const logtypes[] = {
"debug",
};
-void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...)
+void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
{
struct super_block *sb = fs_info->sb;
char lvl[4];
@@ -209,7 +201,7 @@ void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...)
vaf.fmt = fmt;
vaf.va = &args;
- printk("%sBTRFS %s (device %s): %pV", lvl, type, sb->s_id, &vaf);
+ printk("%sBTRFS %s (device %s): %pV\n", lvl, type, sb->s_id, &vaf);
va_end(args);
}
@@ -253,18 +245,24 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root, const char *function,
unsigned int line, int errno)
{
- WARN_ONCE(1, KERN_DEBUG "btrfs: Transaction aborted\n");
+ /*
+ * Report first abort since mount
+ */
+ if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED,
+ &root->fs_info->fs_state)) {
+ WARN(1, KERN_DEBUG "btrfs: Transaction aborted (error %d)\n",
+ errno);
+ }
trans->aborted = errno;
/* Nothing used. The other threads that have joined this
* transaction may be able to continue. */
if (!trans->blocks_used) {
- char nbuf[16];
const char *errstr;
- errstr = btrfs_decode_error(root->fs_info, errno, nbuf);
- btrfs_printk(root->fs_info,
- "%s:%d: Aborting unused transaction(%s).\n",
- function, line, errstr);
+ errstr = btrfs_decode_error(errno);
+ btrfs_warn(root->fs_info,
+ "%s:%d: Aborting unused transaction(%s).",
+ function, line, errstr);
return;
}
ACCESS_ONCE(trans->transaction->aborted) = errno;
@@ -277,7 +275,6 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
void __btrfs_panic(struct btrfs_fs_info *fs_info, const char *function,
unsigned int line, int errno, const char *fmt, ...)
{
- char nbuf[16];
char *s_id = "<unknown>";
const char *errstr;
struct va_format vaf = { .fmt = fmt };
@@ -289,13 +286,13 @@ void __btrfs_panic(struct btrfs_fs_info *fs_info, const char *function,
va_start(args, fmt);
vaf.va = &args;
- errstr = btrfs_decode_error(fs_info, errno, nbuf);
- if (fs_info->mount_opt & BTRFS_MOUNT_PANIC_ON_FATAL_ERROR)
- panic(KERN_CRIT "BTRFS panic (device %s) in %s:%d: %pV (%s)\n",
- s_id, function, line, &vaf, errstr);
+ errstr = btrfs_decode_error(errno);
+ if (fs_info && (fs_info->mount_opt & BTRFS_MOUNT_PANIC_ON_FATAL_ERROR))
+ panic(KERN_CRIT "BTRFS panic (device %s) in %s:%d: %pV (errno=%d %s)\n",
+ s_id, function, line, &vaf, errno, errstr);
- printk(KERN_CRIT "BTRFS panic (device %s) in %s:%d: %pV (%s)\n",
- s_id, function, line, &vaf, errstr);
+ printk(KERN_CRIT "BTRFS panic (device %s) in %s:%d: %pV (errno=%d %s)\n",
+ s_id, function, line, &vaf, errno, errstr);
va_end(args);
/* Caller calls BUG() */
}
@@ -438,6 +435,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
case Opt_compress_force:
case Opt_compress_force_type:
compress_force = true;
+ /* Fallthrough */
case Opt_compress:
case Opt_compress_type:
if (token == Opt_compress ||
@@ -519,7 +517,9 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
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);
printk(KERN_INFO
"btrfs: allocations start at %llu\n",
@@ -648,7 +648,7 @@ out:
*/
static int btrfs_parse_early_options(const char *options, fmode_t flags,
void *holder, char **subvol_name, u64 *subvol_objectid,
- u64 *subvol_rootid, struct btrfs_fs_devices **fs_devices)
+ struct btrfs_fs_devices **fs_devices)
{
substring_t args[MAX_OPT_ARGS];
char *device_name, *opts, *orig, *p;
@@ -691,16 +691,8 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
}
break;
case Opt_subvolrootid:
- intarg = 0;
- error = match_int(&args[0], &intarg);
- if (!error) {
- /* we want the original fs_tree */
- if (!intarg)
- *subvol_rootid =
- BTRFS_FS_TREE_OBJECTID;
- else
- *subvol_rootid = intarg;
- }
+ printk(KERN_WARNING
+ "btrfs: 'subvolrootid' mount option is deprecated and has no effect\n");
break;
case Opt_device:
device_name = match_strdup(&args[0]);
@@ -874,9 +866,9 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
return 0;
}
- btrfs_wait_ordered_extents(root, 0);
+ btrfs_wait_ordered_extents(root, 1);
- trans = btrfs_attach_transaction(root);
+ trans = btrfs_attach_transaction_barrier(root);
if (IS_ERR(trans)) {
/* no transaction, don't bother */
if (PTR_ERR(trans) == -ENOENT)
@@ -1078,7 +1070,6 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
fmode_t mode = FMODE_READ;
char *subvol_name = NULL;
u64 subvol_objectid = 0;
- u64 subvol_rootid = 0;
int error = 0;
if (!(flags & MS_RDONLY))
@@ -1086,7 +1077,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
error = btrfs_parse_early_options(data, mode, fs_type,
&subvol_name, &subvol_objectid,
- &subvol_rootid, &fs_devices);
+ &fs_devices);
if (error) {
kfree(subvol_name);
return ERR_PTR(error);
@@ -1200,6 +1191,41 @@ static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
new_pool_size);
}
+static inline void btrfs_remount_prepare(struct btrfs_fs_info *fs_info)
+{
+ set_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
+}
+
+static inline void btrfs_remount_begin(struct btrfs_fs_info *fs_info,
+ unsigned long old_opts, int flags)
+{
+ if (btrfs_raw_test_opt(old_opts, AUTO_DEFRAG) &&
+ (!btrfs_raw_test_opt(fs_info->mount_opt, AUTO_DEFRAG) ||
+ (flags & MS_RDONLY))) {
+ /* wait for any defraggers to finish */
+ wait_event(fs_info->transaction_wait,
+ (atomic_read(&fs_info->defrag_running) == 0));
+ if (flags & MS_RDONLY)
+ sync_filesystem(fs_info->sb);
+ }
+}
+
+static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
+ unsigned long old_opts)
+{
+ /*
+ * We need cleanup all defragable inodes if the autodefragment is
+ * close or the fs is R/O.
+ */
+ if (btrfs_raw_test_opt(old_opts, AUTO_DEFRAG) &&
+ (!btrfs_raw_test_opt(fs_info->mount_opt, AUTO_DEFRAG) ||
+ (fs_info->sb->s_flags & MS_RDONLY))) {
+ btrfs_cleanup_defrag_inodes(fs_info);
+ }
+
+ clear_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
+}
+
static int btrfs_remount(struct super_block *sb, int *flags, char *data)
{
struct btrfs_fs_info *fs_info = btrfs_sb(sb);
@@ -1213,17 +1239,20 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
unsigned int old_metadata_ratio = fs_info->metadata_ratio;
int ret;
+ btrfs_remount_prepare(fs_info);
+
ret = btrfs_parse_options(root, data);
if (ret) {
ret = -EINVAL;
goto restore;
}
+ btrfs_remount_begin(fs_info, old_opts, *flags);
btrfs_resize_thread_pool(fs_info,
fs_info->thread_pool_size, old_thread_pool_size);
if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
- return 0;
+ goto out;
if (*flags & MS_RDONLY) {
/*
@@ -1278,7 +1307,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
}
sb->s_flags &= ~MS_RDONLY;
}
-
+out:
+ btrfs_remount_cleanup(fs_info, old_opts);
return 0;
restore:
@@ -1289,10 +1319,13 @@ 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;
+ btrfs_remount_cleanup(fs_info, old_opts);
return ret;
}
@@ -1518,6 +1551,7 @@ static struct file_system_type btrfs_fs_type = {
.kill_sb = btrfs_kill_super,
.fs_flags = FS_REQUIRES_DEV,
};
+MODULE_ALIAS_FS("btrfs");
/*
* used by btrfsctl to scan devices when no FS is mounted
@@ -1559,7 +1593,7 @@ static int btrfs_freeze(struct super_block *sb)
struct btrfs_trans_handle *trans;
struct btrfs_root *root = btrfs_sb(sb)->tree_root;
- trans = btrfs_attach_transaction(root);
+ trans = btrfs_attach_transaction_barrier(root);
if (IS_ERR(trans)) {
/* no transaction, don't bother */
if (PTR_ERR(trans) == -ENOENT)
@@ -1684,21 +1718,31 @@ static int __init init_btrfs_fs(void)
if (err)
goto free_delayed_inode;
- err = btrfs_interface_init();
+ err = btrfs_delayed_ref_init();
if (err)
goto free_auto_defrag;
+ err = btrfs_interface_init();
+ if (err)
+ goto free_delayed_ref;
+
err = register_filesystem(&btrfs_fs_type);
if (err)
goto unregister_ioctl;
btrfs_init_lockdep();
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+ btrfs_test_free_space_cache();
+#endif
+
printk(KERN_INFO "%s loaded\n", BTRFS_BUILD_VERSION);
return 0;
unregister_ioctl:
btrfs_interface_exit();
+free_delayed_ref:
+ btrfs_delayed_ref_exit();
free_auto_defrag:
btrfs_auto_defrag_exit();
free_delayed_inode:
@@ -1720,6 +1764,7 @@ free_compress:
static void __exit exit_btrfs_fs(void)
{
btrfs_destroy_cachep();
+ btrfs_delayed_ref_exit();
btrfs_auto_defrag_exit();
btrfs_delayed_inode_exit();
ordered_data_exit();
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index daac9ae6d73..5b326cd60a4 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -21,7 +21,6 @@
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
-#include <linux/module.h>
#include <linux/kobject.h>
#include "ctree.h"
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index fc03aa60b68..0544587d74f 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -34,13 +34,12 @@
#define BTRFS_ROOT_TRANS_TAG 0
-void put_transaction(struct btrfs_transaction *transaction)
+static void put_transaction(struct btrfs_transaction *transaction)
{
WARN_ON(atomic_read(&transaction->use_count) == 0);
if (atomic_dec_and_test(&transaction->use_count)) {
BUG_ON(!list_empty(&transaction->list));
WARN_ON(transaction->delayed_refs.root.rb_node);
- memset(transaction, 0, sizeof(*transaction));
kmem_cache_free(btrfs_transaction_cachep, transaction);
}
}
@@ -51,6 +50,14 @@ static noinline void switch_commit_root(struct btrfs_root *root)
root->commit_root = btrfs_root_node(root);
}
+static inline int can_join_transaction(struct btrfs_transaction *trans,
+ int type)
+{
+ return !(trans->in_commit &&
+ type != TRANS_JOIN &&
+ type != TRANS_JOIN_NOLOCK);
+}
+
/*
* either allocate a new transaction or hop into the existing one
*/
@@ -62,7 +69,7 @@ static noinline int join_transaction(struct btrfs_root *root, int type)
spin_lock(&fs_info->trans_lock);
loop:
/* The file system has been taken offline. No new transactions. */
- if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+ if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
spin_unlock(&fs_info->trans_lock);
return -EROFS;
}
@@ -86,6 +93,10 @@ loop:
spin_unlock(&fs_info->trans_lock);
return cur_trans->aborted;
}
+ if (!can_join_transaction(cur_trans, type)) {
+ spin_unlock(&fs_info->trans_lock);
+ return -EBUSY;
+ }
atomic_inc(&cur_trans->use_count);
atomic_inc(&cur_trans->num_writers);
cur_trans->num_joined++;
@@ -112,9 +123,8 @@ loop:
* to redo the trans_no_join checks above
*/
kmem_cache_free(btrfs_transaction_cachep, cur_trans);
- cur_trans = fs_info->running_transaction;
goto loop;
- } else if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+ } 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);
return -EROFS;
@@ -152,12 +162,16 @@ loop:
if (!RB_EMPTY_ROOT(&fs_info->tree_mod_log))
WARN(1, KERN_ERR "btrfs: tree_mod_log rb tree not empty when "
"creating a fresh transaction\n");
- atomic_set(&fs_info->tree_mod_seq, 0);
+ atomic64_set(&fs_info->tree_mod_seq, 0);
spin_lock_init(&cur_trans->commit_lock);
spin_lock_init(&cur_trans->delayed_refs.lock);
+ atomic_set(&cur_trans->delayed_refs.procs_running_refs, 0);
+ atomic_set(&cur_trans->delayed_refs.ref_seq, 0);
+ init_waitqueue_head(&cur_trans->delayed_refs.wait);
INIT_LIST_HEAD(&cur_trans->pending_snapshots);
+ INIT_LIST_HEAD(&cur_trans->ordered_operations);
list_add_tail(&cur_trans->list, &fs_info->trans_list);
extent_io_tree_init(&cur_trans->dirty_pages,
fs_info->btree_inode->i_mapping);
@@ -302,7 +316,7 @@ start_transaction(struct btrfs_root *root, u64 num_items, int type,
int ret;
u64 qgroup_reserved = 0;
- if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
+ if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
return ERR_PTR(-EROFS);
if (current->journal_info) {
@@ -360,8 +374,11 @@ again:
do {
ret = join_transaction(root, type);
- if (ret == -EBUSY)
+ if (ret == -EBUSY) {
wait_current_trans(root);
+ if (unlikely(type == TRANS_ATTACH))
+ ret = -ENOENT;
+ }
} while (ret == -EBUSY);
if (ret < 0) {
@@ -383,9 +400,10 @@ again:
h->block_rsv = NULL;
h->orig_rsv = NULL;
h->aborted = 0;
- h->qgroup_reserved = qgroup_reserved;
+ h->qgroup_reserved = 0;
h->delayed_ref_elem.seq = 0;
h->type = type;
+ h->allocating_chunk = false;
INIT_LIST_HEAD(&h->qgroup_ref_list);
INIT_LIST_HEAD(&h->new_bgs);
@@ -401,6 +419,7 @@ again:
h->block_rsv = &root->fs_info->trans_block_rsv;
h->bytes_reserved = num_bytes;
}
+ h->qgroup_reserved = qgroup_reserved;
got_it:
btrfs_record_root_in_trans(h, root);
@@ -452,11 +471,43 @@ struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root
return start_transaction(root, 0, TRANS_USERSPACE, 0);
}
+/*
+ * btrfs_attach_transaction() - catch the running transaction
+ *
+ * It is used when we want to commit the current the transaction, but
+ * don't want to start a new one.
+ *
+ * Note: If this function return -ENOENT, it just means there is no
+ * running transaction. But it is possible that the inactive transaction
+ * is still in the memory, not fully on disk. If you hope there is no
+ * inactive transaction in the fs when -ENOENT is returned, you should
+ * invoke
+ * btrfs_attach_transaction_barrier()
+ */
struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root)
{
return start_transaction(root, 0, TRANS_ATTACH, 0);
}
+/*
+ * btrfs_attach_transaction() - catch the running transaction
+ *
+ * It is similar to the above function, the differentia is this one
+ * will wait for all the inactive transactions until they fully
+ * complete.
+ */
+struct btrfs_trans_handle *
+btrfs_attach_transaction_barrier(struct btrfs_root *root)
+{
+ struct btrfs_trans_handle *trans;
+
+ trans = start_transaction(root, 0, TRANS_ATTACH, 0);
+ if (IS_ERR(trans) && PTR_ERR(trans) == -ENOENT)
+ btrfs_wait_for_commit(root, 0);
+
+ return trans;
+}
+
/* wait for a transaction commit to be fully complete */
static noinline void wait_for_commit(struct btrfs_root *root,
struct btrfs_transaction *commit)
@@ -574,21 +625,20 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
btrfs_trans_release_metadata(trans, root);
trans->block_rsv = NULL;
- /*
- * the same root has to be passed to start_transaction and
- * end_transaction. Subvolume quota depends on this.
- */
- WARN_ON(trans->root != root);
if (trans->qgroup_reserved) {
- btrfs_qgroup_free(root, trans->qgroup_reserved);
+ /*
+ * the same root has to be passed here between start_transaction
+ * and end_transaction. Subvolume quota depends on this.
+ */
+ btrfs_qgroup_free(trans->root, trans->qgroup_reserved);
trans->qgroup_reserved = 0;
}
if (!list_empty(&trans->new_bgs))
btrfs_create_pending_block_groups(trans, root);
- while (count < 2) {
+ while (count < 1) {
unsigned long cur = trans->delayed_ref_updates;
trans->delayed_ref_updates = 0;
if (cur &&
@@ -600,6 +650,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
}
count++;
}
+
btrfs_trans_release_metadata(trans, root);
trans->block_rsv = NULL;
@@ -645,12 +696,10 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
btrfs_run_delayed_iputs(root);
if (trans->aborted ||
- root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+ test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
err = -EIO;
- }
assert_qgroups_uptodate(trans);
- memset(trans, 0, sizeof(*trans));
kmem_cache_free(btrfs_trans_handle_cachep, trans);
return err;
}
@@ -658,23 +707,13 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
int btrfs_end_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
- int ret;
-
- ret = __btrfs_end_transaction(trans, root, 0);
- if (ret)
- return ret;
- return 0;
+ return __btrfs_end_transaction(trans, root, 0);
}
int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
- int ret;
-
- ret = __btrfs_end_transaction(trans, root, 1);
- if (ret)
- return ret;
- return 0;
+ return __btrfs_end_transaction(trans, root, 1);
}
int btrfs_end_transaction_dmeta(struct btrfs_trans_handle *trans,
@@ -697,7 +736,9 @@ int btrfs_write_marked_extents(struct btrfs_root *root,
struct extent_state *cached_state = NULL;
u64 start = 0;
u64 end;
+ struct blk_plug plug;
+ blk_start_plug(&plug);
while (!find_first_extent_bit(dirty_pages, start, &start, &end,
mark, &cached_state)) {
convert_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT,
@@ -711,6 +752,7 @@ int btrfs_write_marked_extents(struct btrfs_root *root,
}
if (err)
werr = err;
+ blk_finish_plug(&plug);
return werr;
}
@@ -896,7 +938,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
int btrfs_add_dead_root(struct btrfs_root *root)
{
spin_lock(&root->fs_info->trans_lock);
- list_add(&root->root_list, &root->fs_info->dead_roots);
+ list_add_tail(&root->root_list, &root->fs_info->dead_roots);
spin_unlock(&root->fs_info->trans_lock);
return 0;
}
@@ -961,10 +1003,10 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans,
}
/*
- * defrag a given btree. If cacheonly == 1, this won't read from the disk,
- * otherwise every leaf in the btree is read and defragged.
+ * defrag a given btree.
+ * Every leaf in the btree is read and defragged.
*/
-int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
+int btrfs_defrag_root(struct btrfs_root *root)
{
struct btrfs_fs_info *info = root->fs_info;
struct btrfs_trans_handle *trans;
@@ -978,7 +1020,7 @@ int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
if (IS_ERR(trans))
return PTR_ERR(trans);
- ret = btrfs_defrag_leaves(trans, root, cacheonly);
+ ret = btrfs_defrag_leaves(trans, root);
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(info->tree_root);
@@ -986,6 +1028,12 @@ int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
if (btrfs_fs_closing(root->fs_info) || ret != -EAGAIN)
break;
+
+ if (btrfs_defrag_cancelled(root->fs_info)) {
+ printk(KERN_DEBUG "btrfs: defrag_root cancelled\n");
+ ret = -EAGAIN;
+ break;
+ }
}
root->defrag_running = 0;
return ret;
@@ -993,7 +1041,12 @@ int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
/*
* new snapshots need to be created at a very specific time in the
- * transaction commit. This does the actual creation
+ * transaction commit. This does the actual creation.
+ *
+ * Note:
+ * If the error which may affect the commitment of the current transaction
+ * happens, we should return the error number. If the error which just affect
+ * the creation of the pending snapshots, just return 0.
*/
static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
@@ -1008,12 +1061,11 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
struct inode *parent_inode;
struct btrfs_path *path;
struct btrfs_dir_item *dir_item;
- struct dentry *parent;
struct dentry *dentry;
struct extent_buffer *tmp;
struct extent_buffer *old;
struct timespec cur_time = CURRENT_TIME;
- int ret;
+ int ret = 0;
u64 to_reserve = 0;
u64 index = 0;
u64 objectid;
@@ -1022,40 +1074,36 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
path = btrfs_alloc_path();
if (!path) {
- ret = pending->error = -ENOMEM;
- goto path_alloc_fail;
+ pending->error = -ENOMEM;
+ return 0;
}
new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS);
if (!new_root_item) {
- ret = pending->error = -ENOMEM;
+ pending->error = -ENOMEM;
goto root_item_alloc_fail;
}
- ret = btrfs_find_free_objectid(tree_root, &objectid);
- if (ret) {
- pending->error = ret;
+ pending->error = btrfs_find_free_objectid(tree_root, &objectid);
+ if (pending->error)
goto no_free_objectid;
- }
btrfs_reloc_pre_snapshot(trans, pending, &to_reserve);
if (to_reserve > 0) {
- ret = btrfs_block_rsv_add(root, &pending->block_rsv,
- to_reserve,
- BTRFS_RESERVE_NO_FLUSH);
- if (ret) {
- pending->error = ret;
+ pending->error = btrfs_block_rsv_add(root,
+ &pending->block_rsv,
+ to_reserve,
+ BTRFS_RESERVE_NO_FLUSH);
+ if (pending->error)
goto no_free_objectid;
- }
}
- ret = btrfs_qgroup_inherit(trans, fs_info, root->root_key.objectid,
- objectid, pending->inherit);
- if (ret) {
- pending->error = ret;
+ pending->error = btrfs_qgroup_inherit(trans, fs_info,
+ root->root_key.objectid,
+ objectid, pending->inherit);
+ if (pending->error)
goto no_free_objectid;
- }
key.objectid = objectid;
key.offset = (u64)-1;
@@ -1063,10 +1111,10 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
rsv = trans->block_rsv;
trans->block_rsv = &pending->block_rsv;
+ trans->bytes_reserved = trans->block_rsv->reserved;
dentry = pending->dentry;
- parent = dget_parent(dentry);
- parent_inode = parent->d_inode;
+ parent_inode = pending->dir;
parent_root = BTRFS_I(parent_inode)->root;
record_root_in_trans(trans, parent_root);
@@ -1083,7 +1131,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
dentry->d_name.len, 0);
if (dir_item != NULL && !IS_ERR(dir_item)) {
pending->error = -EEXIST;
- goto fail;
+ goto dir_item_existed;
} else if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
btrfs_abort_transaction(trans, root, ret);
@@ -1121,13 +1169,17 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
memcpy(new_root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE);
memcpy(new_root_item->parent_uuid, root->root_item.uuid,
BTRFS_UUID_SIZE);
+ if (!(root_flags & BTRFS_ROOT_SUBVOL_RDONLY)) {
+ memset(new_root_item->received_uuid, 0,
+ sizeof(new_root_item->received_uuid));
+ memset(&new_root_item->stime, 0, sizeof(new_root_item->stime));
+ memset(&new_root_item->rtime, 0, sizeof(new_root_item->rtime));
+ btrfs_set_root_stransid(new_root_item, 0);
+ btrfs_set_root_rtransid(new_root_item, 0);
+ }
new_root_item->otime.sec = cpu_to_le64(cur_time.tv_sec);
new_root_item->otime.nsec = cpu_to_le32(cur_time.tv_nsec);
btrfs_set_root_otransid(new_root_item, trans->transid);
- memset(&new_root_item->stime, 0, sizeof(new_root_item->stime));
- memset(&new_root_item->rtime, 0, sizeof(new_root_item->rtime));
- btrfs_set_root_stransid(new_root_item, 0);
- btrfs_set_root_rtransid(new_root_item, 0);
old = btrfs_lock_root_node(root);
ret = btrfs_cow_block(trans, root, old, NULL, 0, &old);
@@ -1214,14 +1266,14 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
if (ret)
btrfs_abort_transaction(trans, root, ret);
fail:
- dput(parent);
+ pending->error = ret;
+dir_item_existed:
trans->block_rsv = rsv;
+ trans->bytes_reserved = 0;
no_free_objectid:
kfree(new_root_item);
root_item_alloc_fail:
btrfs_free_path(path);
-path_alloc_fail:
- btrfs_block_rsv_release(root, &pending->block_rsv, (u64)-1);
return ret;
}
@@ -1231,12 +1283,17 @@ path_alloc_fail:
static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info)
{
- struct btrfs_pending_snapshot *pending;
+ struct btrfs_pending_snapshot *pending, *next;
struct list_head *head = &trans->transaction->pending_snapshots;
+ int ret = 0;
- list_for_each_entry(pending, head, list)
- create_pending_snapshot(trans, fs_info, pending);
- return 0;
+ list_for_each_entry_safe(pending, next, head, list) {
+ list_del(&pending->list);
+ ret = create_pending_snapshot(trans, fs_info, pending);
+ if (ret)
+ break;
+ }
+ return ret;
}
static void update_super_roots(struct btrfs_root *root)
@@ -1307,13 +1364,13 @@ static void wait_current_trans_commit_start_and_unblock(struct btrfs_root *root,
struct btrfs_async_commit {
struct btrfs_trans_handle *newtrans;
struct btrfs_root *root;
- struct delayed_work work;
+ struct work_struct work;
};
static void do_async_commit(struct work_struct *work)
{
struct btrfs_async_commit *ac =
- container_of(work, struct btrfs_async_commit, work.work);
+ container_of(work, struct btrfs_async_commit, work);
/*
* We've got freeze protection passed with the transaction.
@@ -1341,7 +1398,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
if (!ac)
return -ENOMEM;
- INIT_DELAYED_WORK(&ac->work, do_async_commit);
+ INIT_WORK(&ac->work, do_async_commit);
ac->root = root;
ac->newtrans = btrfs_join_transaction(root);
if (IS_ERR(ac->newtrans)) {
@@ -1365,7 +1422,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
&root->fs_info->sb->s_writers.lock_map[SB_FREEZE_FS-1],
1, _THIS_IP_);
- schedule_delayed_work(&ac->work, 0);
+ schedule_work(&ac->work);
/* wait for transaction to start and unblock */
if (wait_for_unblock)
@@ -1385,16 +1442,29 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root, int err)
{
struct btrfs_transaction *cur_trans = trans->transaction;
+ DEFINE_WAIT(wait);
WARN_ON(trans->use_count > 1);
btrfs_abort_transaction(trans, root, err);
spin_lock(&root->fs_info->trans_lock);
+
+ if (list_empty(&cur_trans->list)) {
+ spin_unlock(&root->fs_info->trans_lock);
+ btrfs_end_transaction(trans, root);
+ return;
+ }
+
list_del_init(&cur_trans->list);
if (cur_trans == root->fs_info->running_transaction) {
+ root->fs_info->trans_no_join = 1;
+ spin_unlock(&root->fs_info->trans_lock);
+ wait_event(cur_trans->writer_wait,
+ atomic_read(&cur_trans->num_writers) == 1);
+
+ spin_lock(&root->fs_info->trans_lock);
root->fs_info->running_transaction = NULL;
- root->fs_info->trans_no_join = 0;
}
spin_unlock(&root->fs_info->trans_lock);
@@ -1411,6 +1481,10 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
current->journal_info = NULL;
kmem_cache_free(btrfs_trans_handle_cachep, trans);
+
+ spin_lock(&root->fs_info->trans_lock);
+ root->fs_info->trans_no_join = 0;
+ spin_unlock(&root->fs_info->trans_lock);
}
static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
@@ -1428,7 +1502,9 @@ static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
}
if (flush_on_commit || snap_pending) {
- btrfs_start_delalloc_inodes(root, 1);
+ ret = btrfs_start_delalloc_inodes(root, 1);
+ if (ret)
+ return ret;
btrfs_wait_ordered_extents(root, 1);
}
@@ -1450,9 +1526,9 @@ static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
* it here and no for sure that nothing new will be added
* to the list
*/
- btrfs_run_ordered_operations(root, 1);
+ ret = btrfs_run_ordered_operations(trans, root, 1);
- return 0;
+ return ret;
}
/*
@@ -1473,27 +1549,35 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
int should_grow = 0;
unsigned long now = get_seconds();
- ret = btrfs_run_ordered_operations(root, 0);
+ ret = btrfs_run_ordered_operations(trans, root, 0);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
- goto cleanup_transaction;
+ btrfs_end_transaction(trans, root);
+ return ret;
}
/* Stop the commit early if ->aborted is set */
if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
ret = cur_trans->aborted;
- goto cleanup_transaction;
+ btrfs_end_transaction(trans, root);
+ return ret;
}
/* make a pass through all the delayed refs we have so far
* any runnings procs may add more while we are here
*/
ret = btrfs_run_delayed_refs(trans, root, 0);
- if (ret)
- goto cleanup_transaction;
+ if (ret) {
+ btrfs_end_transaction(trans, root);
+ return ret;
+ }
btrfs_trans_release_metadata(trans, root);
trans->block_rsv = NULL;
+ if (trans->qgroup_reserved) {
+ btrfs_qgroup_free(root, trans->qgroup_reserved);
+ trans->qgroup_reserved = 0;
+ }
cur_trans = trans->transaction;
@@ -1507,8 +1591,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
btrfs_create_pending_block_groups(trans, root);
ret = btrfs_run_delayed_refs(trans, root, 0);
- if (ret)
- goto cleanup_transaction;
+ if (ret) {
+ btrfs_end_transaction(trans, root);
+ return ret;
+ }
spin_lock(&cur_trans->commit_lock);
if (cur_trans->in_commit) {
@@ -1720,7 +1806,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
ret = btrfs_write_and_wait_transaction(trans, root);
if (ret) {
btrfs_error(root->fs_info, ret,
- "Error while writing out transaction.");
+ "Error while writing out transaction");
mutex_unlock(&root->fs_info->tree_log_mutex);
goto cleanup_transaction;
}
@@ -1772,8 +1858,11 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
cleanup_transaction:
btrfs_trans_release_metadata(trans, root);
trans->block_rsv = NULL;
- btrfs_printk(root->fs_info, "Skipping commit of aborted transaction.\n");
-// WARN_ON(1);
+ if (trans->qgroup_reserved) {
+ btrfs_qgroup_free(root, trans->qgroup_reserved);
+ trans->qgroup_reserved = 0;
+ }
+ btrfs_warn(root->fs_info, "Skipping commit of aborted transaction.");
if (current->journal_info == trans)
current->journal_info = NULL;
cleanup_transaction(trans, root, ret);
@@ -1782,31 +1871,49 @@ cleanup_transaction:
}
/*
- * interface function to delete all the snapshots we have scheduled for deletion
+ * return < 0 if error
+ * 0 if there are no more dead_roots at the time of call
+ * 1 there are more to be processed, call me again
+ *
+ * The return value indicates there are certainly more snapshots to delete, but
+ * if there comes a new one during processing, it may return 0. We don't mind,
+ * because btrfs_commit_super will poke cleaner thread and it will process it a
+ * few seconds later.
*/
-int btrfs_clean_old_snapshots(struct btrfs_root *root)
+int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root)
{
- LIST_HEAD(list);
+ int ret;
struct btrfs_fs_info *fs_info = root->fs_info;
+ if (fs_info->sb->s_flags & MS_RDONLY) {
+ pr_debug("btrfs: cleaner called for RO fs!\n");
+ return 0;
+ }
+
spin_lock(&fs_info->trans_lock);
- list_splice_init(&fs_info->dead_roots, &list);
+ if (list_empty(&fs_info->dead_roots)) {
+ spin_unlock(&fs_info->trans_lock);
+ return 0;
+ }
+ root = list_first_entry(&fs_info->dead_roots,
+ struct btrfs_root, root_list);
+ list_del(&root->root_list);
spin_unlock(&fs_info->trans_lock);
- while (!list_empty(&list)) {
- int ret;
-
- root = list_entry(list.next, struct btrfs_root, root_list);
- list_del(&root->root_list);
+ pr_debug("btrfs: cleaner removing %llu\n",
+ (unsigned long long)root->objectid);
- btrfs_kill_all_delayed_nodes(root);
+ btrfs_kill_all_delayed_nodes(root);
- if (btrfs_header_backref_rev(root->node) <
- BTRFS_MIXED_BACKREF_REV)
- ret = btrfs_drop_snapshot(root, NULL, 0, 0);
- else
- ret =btrfs_drop_snapshot(root, NULL, 1, 0);
- BUG_ON(ret < 0);
- }
- return 0;
+ if (btrfs_header_backref_rev(root->node) <
+ BTRFS_MIXED_BACKREF_REV)
+ ret = btrfs_drop_snapshot(root, NULL, 0, 0);
+ else
+ ret = btrfs_drop_snapshot(root, NULL, 1, 0);
+ /*
+ * If we encounter a transaction abort during snapshot cleaning, we
+ * don't want to crash here
+ */
+ BUG_ON(ret < 0 && ret != -EAGAIN && ret != -EROFS);
+ return 1;
}
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 0e8aa1e6c28..24c97335a59 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -43,6 +43,7 @@ struct btrfs_transaction {
wait_queue_head_t writer_wait;
wait_queue_head_t commit_wait;
struct list_head pending_snapshots;
+ struct list_head ordered_operations;
struct btrfs_delayed_ref_root delayed_refs;
int aborted;
};
@@ -68,6 +69,7 @@ struct btrfs_trans_handle {
struct btrfs_block_rsv *orig_rsv;
short aborted;
short adding_csums;
+ bool allocating_chunk;
enum btrfs_trans_type type;
/*
* this root is only needed to validate that the root passed to
@@ -82,11 +84,13 @@ struct btrfs_trans_handle {
struct btrfs_pending_snapshot {
struct dentry *dentry;
+ struct inode *dir;
struct btrfs_root *root;
struct btrfs_root *snap;
struct btrfs_qgroup_inherit *inherit;
/* block reservation for the operation */
struct btrfs_block_rsv block_rsv;
+ u64 qgroup_reserved;
/* extra metadata reseration for relocation */
int error;
bool readonly;
@@ -110,14 +114,16 @@ struct btrfs_trans_handle *btrfs_start_transaction_lflush(
struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root);
+struct btrfs_trans_handle *btrfs_attach_transaction_barrier(
+ struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root);
int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid);
int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_add_dead_root(struct btrfs_root *root);
-int btrfs_defrag_root(struct btrfs_root *root, int cacheonly);
-int btrfs_clean_old_snapshots(struct btrfs_root *root);
+int btrfs_defrag_root(struct btrfs_root *root);
+int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root);
int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
@@ -140,5 +146,4 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
struct extent_io_tree *dirty_pages, int mark);
int btrfs_transaction_blocked(struct btrfs_fs_info *info);
int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
-void put_transaction(struct btrfs_transaction *transaction);
#endif
diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c
index 3b580ee8ab1..94e05c1f118 100644
--- a/fs/btrfs/tree-defrag.c
+++ b/fs/btrfs/tree-defrag.c
@@ -23,13 +23,14 @@
#include "transaction.h"
#include "locking.h"
-/* defrag all the leaves in a given btree. If cache_only == 1, don't read
- * things from disk, otherwise read all the leaves and try to get key order to
+/*
+ * Defrag all the leaves in a given btree.
+ * Read all the leaves and try to get key order to
* better reflect disk order
*/
int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, int cache_only)
+ struct btrfs_root *root)
{
struct btrfs_path *path = NULL;
struct btrfs_key key;
@@ -41,9 +42,6 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
u64 last_ret = 0;
u64 min_trans = 0;
- if (cache_only)
- goto out;
-
if (root->fs_info->extent_root == root) {
/*
* there's recursion here right now in the tree locking,
@@ -86,11 +84,8 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
}
path->keep_locks = 1;
- if (cache_only)
- min_trans = root->defrag_trans_start;
- ret = btrfs_search_forward(root, &key, NULL, path,
- cache_only, min_trans);
+ ret = btrfs_search_forward(root, &key, NULL, path, min_trans);
if (ret < 0)
goto out;
if (ret > 0) {
@@ -109,11 +104,11 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
goto out;
}
path->slots[1] = btrfs_header_nritems(path->nodes[1]);
- next_key_ret = btrfs_find_next_key(root, path, &key, 1, cache_only,
+ next_key_ret = btrfs_find_next_key(root, path, &key, 1,
min_trans);
ret = btrfs_realloc_node(trans, root,
path->nodes[1], 0,
- cache_only, &last_ret,
+ &last_ret,
&root->defrag_progress);
if (ret) {
WARN_ON(ret == -EAGAIN);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 9027bb1e746..c276ac9a0ec 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -277,18 +277,19 @@ static int process_one_buffer(struct btrfs_root *log,
struct extent_buffer *eb,
struct walk_control *wc, u64 gen)
{
+ int ret = 0;
+
if (wc->pin)
- btrfs_pin_extent_for_log_replay(wc->trans,
- log->fs_info->extent_root,
- eb->start, eb->len);
+ ret = btrfs_pin_extent_for_log_replay(log->fs_info->extent_root,
+ eb->start, eb->len);
- if (btrfs_buffer_uptodate(eb, gen, 0)) {
+ if (!ret && btrfs_buffer_uptodate(eb, gen, 0)) {
if (wc->write)
btrfs_write_tree_block(eb);
if (wc->wait)
btrfs_wait_tree_block_writeback(eb);
}
- return 0;
+ return ret;
}
/*
@@ -318,6 +319,7 @@ static noinline int overwrite_item(struct btrfs_trans_handle *trans,
unsigned long src_ptr;
unsigned long dst_ptr;
int overwrite_root = 0;
+ bool inode_item = key->type == BTRFS_INODE_ITEM_KEY;
if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
overwrite_root = 1;
@@ -327,6 +329,9 @@ static noinline int overwrite_item(struct btrfs_trans_handle *trans,
/* look for the key in the destination tree */
ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
+ if (ret < 0)
+ return ret;
+
if (ret == 0) {
char *src_copy;
char *dst_copy;
@@ -368,6 +373,30 @@ static noinline int overwrite_item(struct btrfs_trans_handle *trans,
return 0;
}
+ /*
+ * We need to load the old nbytes into the inode so when we
+ * replay the extents we've logged we get the right nbytes.
+ */
+ if (inode_item) {
+ struct btrfs_inode_item *item;
+ u64 nbytes;
+
+ item = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_inode_item);
+ nbytes = btrfs_inode_nbytes(path->nodes[0], item);
+ item = btrfs_item_ptr(eb, slot,
+ struct btrfs_inode_item);
+ btrfs_set_inode_nbytes(eb, item, nbytes);
+ }
+ } else if (inode_item) {
+ struct btrfs_inode_item *item;
+
+ /*
+ * New inode, set nbytes to 0 so that the nbytes comes out
+ * properly when we replay the extents.
+ */
+ item = btrfs_item_ptr(eb, slot, struct btrfs_inode_item);
+ btrfs_set_inode_nbytes(eb, item, 0);
}
insert:
btrfs_release_path(path);
@@ -381,9 +410,9 @@ insert:
found_size = btrfs_item_size_nr(path->nodes[0],
path->slots[0]);
if (found_size > item_size)
- btrfs_truncate_item(trans, root, path, item_size, 1);
+ btrfs_truncate_item(root, path, item_size, 1);
else if (found_size < item_size)
- btrfs_extend_item(trans, root, path,
+ btrfs_extend_item(root, path,
item_size - found_size);
} else if (ret) {
return ret;
@@ -485,10 +514,9 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
struct btrfs_key *key)
{
int found_type;
- u64 mask = root->sectorsize - 1;
u64 extent_end;
u64 start = key->offset;
- u64 saved_nbytes;
+ u64 nbytes = 0;
struct btrfs_file_extent_item *item;
struct inode *inode = NULL;
unsigned long size;
@@ -498,11 +526,20 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
found_type = btrfs_file_extent_type(eb, item);
if (found_type == BTRFS_FILE_EXTENT_REG ||
- found_type == BTRFS_FILE_EXTENT_PREALLOC)
- extent_end = start + btrfs_file_extent_num_bytes(eb, item);
- else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
+ found_type == BTRFS_FILE_EXTENT_PREALLOC) {
+ nbytes = btrfs_file_extent_num_bytes(eb, item);
+ extent_end = start + nbytes;
+
+ /*
+ * We don't add to the inodes nbytes if we are prealloc or a
+ * hole.
+ */
+ if (btrfs_file_extent_disk_bytenr(eb, item) == 0)
+ nbytes = 0;
+ } else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
size = btrfs_file_extent_inline_len(eb, item);
- extent_end = (start + size + mask) & ~mask;
+ nbytes = btrfs_file_extent_ram_bytes(eb, item);
+ extent_end = ALIGN(start + size, root->sectorsize);
} else {
ret = 0;
goto out;
@@ -550,10 +587,10 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
- saved_nbytes = inode_get_bytes(inode);
/* drop any overlapping extents */
ret = btrfs_drop_extents(trans, root, inode, start, extent_end, 1);
- BUG_ON(ret);
+ if (ret)
+ goto out;
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -563,7 +600,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
ret = btrfs_insert_empty_item(trans, root, path, key,
sizeof(*item));
- BUG_ON(ret);
+ if (ret)
+ goto out;
dest_offset = btrfs_item_ptr_offset(path->nodes[0],
path->slots[0]);
copy_extent_buffer(path->nodes[0], eb, dest_offset,
@@ -589,7 +627,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
ins.objectid, ins.offset,
0, root->root_key.objectid,
key->objectid, offset, 0);
- BUG_ON(ret);
+ if (ret)
+ goto out;
} else {
/*
* insert the extent pointer in the extent
@@ -598,7 +637,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
ret = btrfs_alloc_logged_file_extent(trans,
root, root->root_key.objectid,
key->objectid, offset, &ins);
- BUG_ON(ret);
+ if (ret)
+ goto out;
}
btrfs_release_path(path);
@@ -615,29 +655,33 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
ret = btrfs_lookup_csums_range(root->log_root,
csum_start, csum_end - 1,
&ordered_sums, 0);
- BUG_ON(ret);
+ if (ret)
+ goto out;
while (!list_empty(&ordered_sums)) {
struct btrfs_ordered_sum *sums;
sums = list_entry(ordered_sums.next,
struct btrfs_ordered_sum,
list);
- ret = btrfs_csum_file_blocks(trans,
+ if (!ret)
+ ret = btrfs_csum_file_blocks(trans,
root->fs_info->csum_root,
sums);
- BUG_ON(ret);
list_del(&sums->list);
kfree(sums);
}
+ if (ret)
+ goto out;
} else {
btrfs_release_path(path);
}
} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
/* inline extents are easy, we just overwrite them */
ret = overwrite_item(trans, root, path, eb, slot, key);
- BUG_ON(ret);
+ if (ret)
+ goto out;
}
- inode_set_bytes(inode, saved_nbytes);
+ inode_add_bytes(inode, nbytes);
ret = btrfs_update_inode(trans, root, inode);
out:
if (inode)
@@ -679,20 +723,21 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
inode = read_one_inode(root, location.objectid);
if (!inode) {
- kfree(name);
- return -EIO;
+ ret = -EIO;
+ goto out;
}
ret = link_to_fixup_dir(trans, root, path, location.objectid);
- BUG_ON(ret);
+ if (ret)
+ goto out;
ret = btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
- BUG_ON(ret);
+ if (ret)
+ goto out;
+ btrfs_run_delayed_items(trans, root);
+out:
kfree(name);
-
iput(inode);
-
- btrfs_run_delayed_items(trans, root);
return ret;
}
@@ -845,7 +890,8 @@ again:
victim_name_len = btrfs_inode_ref_name_len(leaf,
victim_ref);
victim_name = kmalloc(victim_name_len, GFP_NOFS);
- BUG_ON(!victim_name);
+ if (!victim_name)
+ return -ENOMEM;
read_extent_buffer(leaf, victim_name,
(unsigned long)(victim_ref + 1),
@@ -861,9 +907,10 @@ again:
ret = btrfs_unlink_inode(trans, root, dir,
inode, victim_name,
victim_name_len);
- BUG_ON(ret);
- btrfs_run_delayed_items(trans, root);
kfree(victim_name);
+ if (ret)
+ return ret;
+ btrfs_run_delayed_items(trans, root);
*search_done = 1;
goto again;
}
@@ -871,7 +918,6 @@ again:
ptr = (unsigned long)(victim_ref + 1) + victim_name_len;
}
- BUG_ON(ret);
/*
* NOTE: we have searched root tree and checked the
@@ -905,6 +951,8 @@ again:
goto next;
victim_name = kmalloc(victim_name_len, GFP_NOFS);
+ if (!victim_name)
+ return -ENOMEM;
read_extent_buffer(leaf, victim_name, (unsigned long)&extref->name,
victim_name_len);
@@ -931,14 +979,16 @@ again:
victim_name_len);
btrfs_run_delayed_items(trans, root);
}
- BUG_ON(ret);
iput(victim_parent);
kfree(victim_name);
+ if (ret)
+ return ret;
*search_done = 1;
goto again;
}
kfree(victim_name);
- BUG_ON(ret);
+ if (ret)
+ return ret;
next:
cur_offset += victim_name_len + sizeof(*extref);
}
@@ -951,7 +1001,8 @@ next:
ref_index, name, namelen, 0);
if (di && !IS_ERR(di)) {
ret = drop_one_dir_item(trans, root, path, dir, di);
- BUG_ON(ret);
+ if (ret)
+ return ret;
}
btrfs_release_path(path);
@@ -960,7 +1011,8 @@ next:
name, namelen, 0);
if (di && !IS_ERR(di)) {
ret = drop_one_dir_item(trans, root, path, dir, di);
- BUG_ON(ret);
+ if (ret)
+ return ret;
}
btrfs_release_path(path);
@@ -1105,15 +1157,19 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
parent_objectid,
ref_index, name, namelen,
&search_done);
- if (ret == 1)
+ if (ret == 1) {
+ ret = 0;
+ goto out;
+ }
+ if (ret)
goto out;
- BUG_ON(ret);
}
/* insert our name */
ret = btrfs_add_link(trans, dir, inode, name, namelen,
0, ref_index);
- BUG_ON(ret);
+ if (ret)
+ goto out;
btrfs_update_inode(trans, root, inode);
}
@@ -1128,13 +1184,11 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
/* finally write the back reference in the inode */
ret = overwrite_item(trans, root, path, eb, slot, key);
- BUG_ON(ret);
-
out:
btrfs_release_path(path);
iput(dir);
iput(inode);
- return 0;
+ return ret;
}
static int insert_orphan_item(struct btrfs_trans_handle *trans,
@@ -1292,10 +1346,10 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
if (S_ISDIR(inode->i_mode)) {
ret = replay_dir_deletes(trans, root, NULL, path,
ino, 1);
- BUG_ON(ret);
+ if (ret)
+ goto out;
}
ret = insert_orphan_item(trans, root, ino);
- BUG_ON(ret);
}
out:
@@ -1340,9 +1394,9 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
return -EIO;
ret = fixup_inode_link_count(trans, root, inode);
- BUG_ON(ret);
-
iput(inode);
+ if (ret)
+ goto out;
/*
* fixup on a directory may create new entries,
@@ -1384,12 +1438,15 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
btrfs_release_path(path);
if (ret == 0) {
- btrfs_inc_nlink(inode);
+ if (!inode->i_nlink)
+ set_nlink(inode, 1);
+ else
+ btrfs_inc_nlink(inode);
ret = btrfs_update_inode(trans, root, inode);
} else if (ret == -EEXIST) {
ret = 0;
} else {
- BUG();
+ BUG(); /* Logic Error */
}
iput(inode);
@@ -1458,7 +1515,7 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
struct inode *dir;
u8 log_type;
int exists;
- int ret;
+ int ret = 0;
dir = read_one_inode(root, key->objectid);
if (!dir)
@@ -1490,7 +1547,9 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
key->offset, name,
name_len, 1);
} else {
- BUG();
+ /* Corruption */
+ ret = -EINVAL;
+ goto out;
}
if (IS_ERR_OR_NULL(dst_di)) {
/* we need a sequence number to insert, so we only
@@ -1518,7 +1577,8 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
goto out;
ret = drop_one_dir_item(trans, root, path, dir, dst_di);
- BUG_ON(ret);
+ if (ret)
+ goto out;
if (key->type == BTRFS_DIR_INDEX_KEY)
goto insert;
@@ -1526,14 +1586,15 @@ out:
btrfs_release_path(path);
kfree(name);
iput(dir);
- return 0;
+ return ret;
insert:
btrfs_release_path(path);
ret = insert_one_name(trans, root, path, key->objectid, key->offset,
name, name_len, log_type, &log_key);
-
- BUG_ON(ret && ret != -ENOENT);
+ if (ret && ret != -ENOENT)
+ goto out;
+ ret = 0;
goto out;
}
@@ -1564,7 +1625,8 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
return -EIO;
name_len = btrfs_dir_name_len(eb, di);
ret = replay_one_name(trans, root, path, eb, di, key);
- BUG_ON(ret);
+ if (ret)
+ return ret;
ptr = (unsigned long)(di + 1);
ptr += name_len;
}
@@ -1725,16 +1787,21 @@ again:
ret = link_to_fixup_dir(trans, root,
path, location.objectid);
- BUG_ON(ret);
+ if (ret) {
+ kfree(name);
+ iput(inode);
+ goto out;
+ }
+
btrfs_inc_nlink(inode);
ret = btrfs_unlink_inode(trans, root, dir, inode,
name, name_len);
- BUG_ON(ret);
-
- btrfs_run_delayed_items(trans, root);
-
+ if (!ret)
+ btrfs_run_delayed_items(trans, root);
kfree(name);
iput(inode);
+ if (ret)
+ goto out;
/* there might still be more names under this key
* check and repeat if required
@@ -1838,7 +1905,8 @@ again:
ret = check_item_in_log(trans, root, log, path,
log_path, dir,
&found_key);
- BUG_ON(ret);
+ if (ret)
+ goto out;
if (found_key.offset == (u64)-1)
break;
dir_key.offset = found_key.offset + 1;
@@ -1915,11 +1983,13 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
if (S_ISDIR(mode)) {
ret = replay_dir_deletes(wc->trans,
root, log, path, key.objectid, 0);
- BUG_ON(ret);
+ if (ret)
+ break;
}
ret = overwrite_item(wc->trans, root, path,
eb, i, &key);
- BUG_ON(ret);
+ if (ret)
+ break;
/* for regular files, make sure corresponding
* orhpan item exist. extents past the new EOF
@@ -1928,12 +1998,14 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
if (S_ISREG(mode)) {
ret = insert_orphan_item(wc->trans, root,
key.objectid);
- BUG_ON(ret);
+ if (ret)
+ break;
}
ret = link_to_fixup_dir(wc->trans, root,
path, key.objectid);
- BUG_ON(ret);
+ if (ret)
+ break;
}
if (wc->stage < LOG_WALK_REPLAY_ALL)
continue;
@@ -1942,28 +2014,35 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
if (key.type == BTRFS_XATTR_ITEM_KEY) {
ret = overwrite_item(wc->trans, root, path,
eb, i, &key);
- BUG_ON(ret);
+ if (ret)
+ break;
} else if (key.type == BTRFS_INODE_REF_KEY) {
ret = add_inode_ref(wc->trans, root, log, path,
eb, i, &key);
- BUG_ON(ret && ret != -ENOENT);
+ if (ret && ret != -ENOENT)
+ break;
+ ret = 0;
} else if (key.type == BTRFS_INODE_EXTREF_KEY) {
ret = add_inode_ref(wc->trans, root, log, path,
eb, i, &key);
- BUG_ON(ret && ret != -ENOENT);
+ if (ret && ret != -ENOENT)
+ break;
+ ret = 0;
} else if (key.type == BTRFS_EXTENT_DATA_KEY) {
ret = replay_one_extent(wc->trans, root, path,
eb, i, &key);
- BUG_ON(ret);
+ if (ret)
+ break;
} else if (key.type == BTRFS_DIR_ITEM_KEY ||
key.type == BTRFS_DIR_INDEX_KEY) {
ret = replay_one_dir_item(wc->trans, root, path,
eb, i, &key);
- BUG_ON(ret);
+ if (ret)
+ break;
}
}
btrfs_free_path(path);
- return 0;
+ return ret;
}
static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
@@ -2008,8 +2087,10 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
if (*level == 1) {
ret = wc->process_func(root, next, wc, ptr_gen);
- if (ret)
+ if (ret) {
+ free_extent_buffer(next);
return ret;
+ }
path->slots[*level]++;
if (wc->free) {
@@ -2029,7 +2110,10 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_and_pin_reserved_extent(root,
bytenr, blocksize);
- BUG_ON(ret); /* -ENOMEM or logic errors */
+ if (ret) {
+ free_extent_buffer(next);
+ return ret;
+ }
}
free_extent_buffer(next);
continue;
@@ -2102,7 +2186,8 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
ret = btrfs_free_and_pin_reserved_extent(root,
path->nodes[*level]->start,
path->nodes[*level]->len);
- BUG_ON(ret);
+ if (ret)
+ return ret;
}
free_extent_buffer(path->nodes[*level]);
path->nodes[*level] = NULL;
@@ -2124,7 +2209,6 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
int wret;
int level;
struct btrfs_path *path;
- int i;
int orig_level;
path = btrfs_alloc_path();
@@ -2176,17 +2260,12 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_and_pin_reserved_extent(log, next->start,
next->len);
- BUG_ON(ret); /* -ENOMEM or logic errors */
+ if (ret)
+ goto out;
}
}
out:
- for (i = 0; i <= orig_level; i++) {
- if (path->nodes[i]) {
- free_extent_buffer(path->nodes[i]);
- path->nodes[i] = NULL;
- }
- }
btrfs_free_path(path);
return ret;
}
@@ -2281,6 +2360,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
unsigned long log_transid = 0;
mutex_lock(&root->log_mutex);
+ log_transid = root->log_transid;
index1 = root->log_transid % 2;
if (atomic_read(&root->log_commit[index1])) {
wait_log_commit(trans, root, root->log_transid);
@@ -2308,11 +2388,11 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
/* bail out if we need to do a full commit */
if (root->fs_info->last_trans_log_full_commit == trans->transid) {
ret = -EAGAIN;
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&root->log_mutex);
goto out;
}
- log_transid = root->log_transid;
if (log_transid % 2 == 0)
mark = EXTENT_DIRTY;
else
@@ -2324,6 +2404,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&root->log_mutex);
goto out;
}
@@ -2363,6 +2444,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
}
root->fs_info->last_trans_log_full_commit = trans->transid;
btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&log_root_tree->log_mutex);
ret = -EAGAIN;
goto out;
@@ -2373,6 +2455,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
wait_log_commit(trans, log_root_tree,
log_root_tree->log_transid);
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&log_root_tree->log_mutex);
ret = 0;
goto out;
@@ -2392,6 +2475,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
*/
if (root->fs_info->last_trans_log_full_commit == trans->transid) {
btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&log_root_tree->log_mutex);
ret = -EAGAIN;
goto out_wake_log_root;
@@ -2402,10 +2486,12 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
EXTENT_DIRTY | EXTENT_NEW);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
+ btrfs_free_logged_extents(log, log_transid);
mutex_unlock(&log_root_tree->log_mutex);
goto out_wake_log_root;
}
btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
+ btrfs_wait_logged_extents(log, log_transid);
btrfs_set_super_log_root(root->fs_info->super_for_commit,
log_root_tree->node->start);
@@ -2461,8 +2547,13 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
.process_func = process_one_buffer
};
- ret = walk_log_tree(trans, log, &wc);
- BUG_ON(ret);
+ if (trans) {
+ ret = walk_log_tree(trans, log, &wc);
+
+ /* I don't think this can happen but just in case */
+ if (ret)
+ btrfs_abort_transaction(trans, log, ret);
+ }
while (1) {
ret = find_first_extent_bit(&log->dirty_log_pages,
@@ -2475,6 +2566,14 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
}
+ /*
+ * We may have short-circuited the log tree with the full commit logic
+ * and left ordered extents on our list, so clear these out to keep us
+ * from leaking inodes and memory.
+ */
+ btrfs_free_logged_extents(log, 0);
+ btrfs_free_logged_extents(log, 1);
+
free_extent_buffer(log->node);
kfree(log);
}
@@ -2561,7 +2660,10 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
if (di) {
ret = btrfs_delete_one_dir_name(trans, log, path, di);
bytes_del += name_len;
- BUG_ON(ret);
+ if (ret) {
+ err = ret;
+ goto fail;
+ }
}
btrfs_release_path(path);
di = btrfs_lookup_dir_index_item(trans, log, path, dir_ino,
@@ -2573,7 +2675,10 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
if (di) {
ret = btrfs_delete_one_dir_name(trans, log, path, di);
bytes_del += name_len;
- BUG_ON(ret);
+ if (ret) {
+ err = ret;
+ goto fail;
+ }
}
/* update the directory size in the log to reflect the names
@@ -2724,7 +2829,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
path->keep_locks = 1;
ret = btrfs_search_forward(root, &min_key, &max_key,
- path, 0, trans->transid);
+ path, trans->transid);
/*
* we didn't find anything from this transaction, see if there
@@ -2912,7 +3017,7 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans,
while (1) {
ret = btrfs_search_slot(trans, log, &key, path, -1, 1);
- BUG_ON(ret == 0);
+ BUG_ON(ret == 0); /* Logic error */
if (ret < 0)
break;
@@ -3115,7 +3220,11 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
log->fs_info->csum_root,
ds + cs, ds + cs + cl - 1,
&ordered_sums, 0);
- BUG_ON(ret);
+ if (ret) {
+ btrfs_release_path(dst_path);
+ kfree(ins_data);
+ return ret;
+ }
}
}
}
@@ -3155,115 +3264,6 @@ static int extent_cmp(void *priv, struct list_head *a, struct list_head *b)
return 0;
}
-static int drop_adjacent_extents(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
- struct extent_map *em,
- struct btrfs_path *path)
-{
- struct btrfs_file_extent_item *fi;
- struct extent_buffer *leaf;
- struct btrfs_key key, new_key;
- struct btrfs_map_token token;
- u64 extent_end;
- u64 extent_offset = 0;
- int extent_type;
- int del_slot = 0;
- int del_nr = 0;
- int ret = 0;
-
- while (1) {
- btrfs_init_map_token(&token);
- leaf = path->nodes[0];
- path->slots[0]++;
- if (path->slots[0] >= btrfs_header_nritems(leaf)) {
- if (del_nr) {
- ret = btrfs_del_items(trans, root, path,
- del_slot, del_nr);
- if (ret)
- return ret;
- del_nr = 0;
- }
-
- ret = btrfs_next_leaf_write(trans, root, path, 1);
- if (ret < 0)
- return ret;
- if (ret > 0)
- return 0;
- leaf = path->nodes[0];
- }
-
- btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
- if (key.objectid != btrfs_ino(inode) ||
- key.type != BTRFS_EXTENT_DATA_KEY ||
- key.offset >= em->start + em->len)
- break;
-
- fi = btrfs_item_ptr(leaf, path->slots[0],
- struct btrfs_file_extent_item);
- extent_type = btrfs_token_file_extent_type(leaf, fi, &token);
- if (extent_type == BTRFS_FILE_EXTENT_REG ||
- extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
- extent_offset = btrfs_token_file_extent_offset(leaf,
- fi, &token);
- extent_end = key.offset +
- btrfs_token_file_extent_num_bytes(leaf, fi,
- &token);
- } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
- extent_end = key.offset +
- btrfs_file_extent_inline_len(leaf, fi);
- } else {
- BUG();
- }
-
- if (extent_end <= em->len + em->start) {
- if (!del_nr) {
- del_slot = path->slots[0];
- }
- del_nr++;
- continue;
- }
-
- /*
- * Ok so we'll ignore previous items if we log a new extent,
- * which can lead to overlapping extents, so if we have an
- * existing extent we want to adjust we _have_ to check the next
- * guy to make sure we even need this extent anymore, this keeps
- * us from panicing in set_item_key_safe.
- */
- if (path->slots[0] < btrfs_header_nritems(leaf) - 1) {
- struct btrfs_key tmp_key;
-
- btrfs_item_key_to_cpu(leaf, &tmp_key,
- path->slots[0] + 1);
- if (tmp_key.objectid == btrfs_ino(inode) &&
- tmp_key.type == BTRFS_EXTENT_DATA_KEY &&
- tmp_key.offset <= em->start + em->len) {
- if (!del_nr)
- del_slot = path->slots[0];
- del_nr++;
- continue;
- }
- }
-
- BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
- memcpy(&new_key, &key, sizeof(new_key));
- new_key.offset = em->start + em->len;
- btrfs_set_item_key_safe(trans, root, path, &new_key);
- extent_offset += em->start + em->len - key.offset;
- btrfs_set_token_file_extent_offset(leaf, fi, extent_offset,
- &token);
- btrfs_set_token_file_extent_num_bytes(leaf, fi, extent_end -
- (em->start + em->len),
- &token);
- btrfs_mark_buffer_dirty(leaf);
- }
-
- if (del_nr)
- ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
-
- return ret;
-}
-
static int log_one_extent(struct btrfs_trans_handle *trans,
struct inode *inode, struct btrfs_root *root,
struct extent_map *em, struct btrfs_path *path)
@@ -3271,31 +3271,38 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *log = root->log_root;
struct btrfs_file_extent_item *fi;
struct extent_buffer *leaf;
+ struct btrfs_ordered_extent *ordered;
struct list_head ordered_sums;
struct btrfs_map_token token;
struct btrfs_key key;
- u64 csum_offset = em->mod_start - em->start;
- u64 csum_len = em->mod_len;
+ u64 mod_start = em->mod_start;
+ u64 mod_len = em->mod_len;
+ u64 csum_offset;
+ u64 csum_len;
u64 extent_offset = em->start - em->orig_start;
u64 block_len;
int ret;
+ int index = log->log_transid % 2;
bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
+ ret = __btrfs_drop_extents(trans, log, inode, path, em->start,
+ em->start + em->len, NULL, 0);
+ if (ret)
+ return ret;
+
INIT_LIST_HEAD(&ordered_sums);
btrfs_init_map_token(&token);
key.objectid = btrfs_ino(inode);
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = em->start;
- path->really_keep_locks = 1;
ret = btrfs_insert_empty_item(trans, log, path, &key, sizeof(*fi));
- if (ret && ret != -EEXIST) {
- path->really_keep_locks = 0;
+ if (ret)
return ret;
- }
leaf = path->nodes[0];
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
+
btrfs_set_token_file_extent_generation(leaf, fi, em->generation,
&token);
if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
@@ -3334,22 +3341,14 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
em->start - em->orig_start,
&token);
btrfs_set_token_file_extent_num_bytes(leaf, fi, em->len, &token);
- btrfs_set_token_file_extent_ram_bytes(leaf, fi, em->len, &token);
+ btrfs_set_token_file_extent_ram_bytes(leaf, fi, em->ram_bytes, &token);
btrfs_set_token_file_extent_compression(leaf, fi, em->compress_type,
&token);
btrfs_set_token_file_extent_encryption(leaf, fi, 0, &token);
btrfs_set_token_file_extent_other_encoding(leaf, fi, 0, &token);
btrfs_mark_buffer_dirty(leaf);
- /*
- * Have to check the extent to the right of us to make sure it doesn't
- * fall in our current range. We're ok if the previous extent is in our
- * range since the recovery stuff will run us in key order and thus just
- * drop the part we overwrote.
- */
- ret = drop_adjacent_extents(trans, log, inode, em, path);
btrfs_release_path(path);
- path->really_keep_locks = 0;
if (ret) {
return ret;
}
@@ -3362,6 +3361,92 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
csum_len = block_len;
}
+ /*
+ * First check and see if our csums are on our outstanding ordered
+ * extents.
+ */
+again:
+ spin_lock_irq(&log->log_extents_lock[index]);
+ list_for_each_entry(ordered, &log->logged_list[index], log_list) {
+ struct btrfs_ordered_sum *sum;
+
+ if (!mod_len)
+ break;
+
+ if (ordered->inode != inode)
+ continue;
+
+ if (ordered->file_offset + ordered->len <= mod_start ||
+ mod_start + mod_len <= ordered->file_offset)
+ continue;
+
+ /*
+ * We are going to copy all the csums on this ordered extent, so
+ * go ahead and adjust mod_start and mod_len in case this
+ * ordered extent has already been logged.
+ */
+ if (ordered->file_offset > mod_start) {
+ if (ordered->file_offset + ordered->len >=
+ mod_start + mod_len)
+ mod_len = ordered->file_offset - mod_start;
+ /*
+ * If we have this case
+ *
+ * |--------- logged extent ---------|
+ * |----- ordered extent ----|
+ *
+ * Just don't mess with mod_start and mod_len, we'll
+ * just end up logging more csums than we need and it
+ * will be ok.
+ */
+ } else {
+ if (ordered->file_offset + ordered->len <
+ mod_start + mod_len) {
+ mod_len = (mod_start + mod_len) -
+ (ordered->file_offset + ordered->len);
+ mod_start = ordered->file_offset +
+ ordered->len;
+ } else {
+ mod_len = 0;
+ }
+ }
+
+ /*
+ * To keep us from looping for the above case of an ordered
+ * extent that falls inside of the logged extent.
+ */
+ if (test_and_set_bit(BTRFS_ORDERED_LOGGED_CSUM,
+ &ordered->flags))
+ continue;
+ atomic_inc(&ordered->refs);
+ spin_unlock_irq(&log->log_extents_lock[index]);
+ /*
+ * we've dropped the lock, we must either break or
+ * start over after this.
+ */
+
+ wait_event(ordered->wait, ordered->csum_bytes_left == 0);
+
+ list_for_each_entry(sum, &ordered->list, list) {
+ ret = btrfs_csum_file_blocks(trans, log, sum);
+ if (ret) {
+ btrfs_put_ordered_extent(ordered);
+ goto unlocked;
+ }
+ }
+ btrfs_put_ordered_extent(ordered);
+ goto again;
+
+ }
+ spin_unlock_irq(&log->log_extents_lock[index]);
+unlocked:
+
+ if (!mod_len || ret)
+ return ret;
+
+ csum_offset = mod_start - em->start;
+ csum_len = mod_len;
+
/* block start is already adjusted for the file extent offset. */
ret = btrfs_lookup_csums_range(log->fs_info->csum_root,
em->block_start + csum_offset,
@@ -3393,6 +3478,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree;
u64 test_gen;
int ret = 0;
+ int num = 0;
INIT_LIST_HEAD(&extents);
@@ -3401,16 +3487,31 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
list_for_each_entry_safe(em, n, &tree->modified_extents, list) {
list_del_init(&em->list);
+
+ /*
+ * Just an arbitrary number, this can be really CPU intensive
+ * once we start getting a lot of extents, and really once we
+ * have a bunch of extents we just want to commit since it will
+ * be faster.
+ */
+ if (++num > 32768) {
+ list_del_init(&tree->modified_extents);
+ ret = -EFBIG;
+ goto process;
+ }
+
if (em->generation <= test_gen)
continue;
/* Need a ref to keep it from getting evicted from cache */
atomic_inc(&em->refs);
set_bit(EXTENT_FLAG_LOGGING, &em->flags);
list_add_tail(&em->list, &extents);
+ num++;
}
list_sort(NULL, &extents, extent_cmp);
+process:
while (!list_empty(&extents)) {
em = list_entry(extents.next, struct extent_map, list);
@@ -3472,8 +3573,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
bool fast_search = false;
u64 ino = btrfs_ino(inode);
- log = root->log_root;
-
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
@@ -3513,6 +3612,8 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
mutex_lock(&BTRFS_I(inode)->log_mutex);
+ btrfs_get_logged_extents(log, inode);
+
/*
* a brute force approach to making sure we get the most uptodate
* copies of everything.
@@ -3558,7 +3659,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
while (1) {
ins_nr = 0;
ret = btrfs_search_forward(root, &min_key, &max_key,
- path, 0, trans->transid);
+ path, trans->transid);
if (ret != 0)
break;
again:
@@ -3656,6 +3757,8 @@ log_extents:
BTRFS_I(inode)->logged_trans = trans->transid;
BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans;
out_unlock:
+ if (err)
+ btrfs_free_logged_extents(log, log->log_transid);
mutex_unlock(&BTRFS_I(inode)->log_mutex);
btrfs_free_path(path);
@@ -3736,9 +3839,9 @@ out:
* only logging is done of any parent directories that are older than
* the last committed transaction
*/
-int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
- struct dentry *parent, int exists_only)
+static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, struct inode *inode,
+ struct dentry *parent, int exists_only)
{
int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
struct super_block *sb;
@@ -3822,7 +3925,6 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
end_trans:
dput(old_parent);
if (ret < 0) {
- WARN_ON(ret != -ENOSPC);
root->fs_info->last_trans_log_full_commit = trans->transid;
ret = 1;
}
@@ -3930,6 +4032,9 @@ again:
wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
if (IS_ERR(wc.replay_dest)) {
ret = PTR_ERR(wc.replay_dest);
+ free_extent_buffer(log->node);
+ free_extent_buffer(log->commit_root);
+ kfree(log);
btrfs_error(fs_info, ret, "Couldn't read target root "
"for tree log recovery.");
goto error;
@@ -3938,12 +4043,10 @@ again:
wc.replay_dest->log_root = log;
btrfs_record_root_in_trans(trans, wc.replay_dest);
ret = walk_log_tree(trans, log, &wc);
- BUG_ON(ret);
- if (wc.stage == LOG_WALK_REPLAY_ALL) {
+ if (!ret && wc.stage == LOG_WALK_REPLAY_ALL) {
ret = fixup_inode_link_counts(trans, wc.replay_dest,
path);
- BUG_ON(ret);
}
key.offset = found_key.offset - 1;
@@ -3952,6 +4055,9 @@ again:
free_extent_buffer(log->commit_root);
kfree(log);
+ if (ret)
+ goto error;
+
if (found_key.offset == 0)
break;
}
@@ -3972,17 +4078,20 @@ again:
btrfs_free_path(path);
+ /* step 4: commit the transaction, which also unpins the blocks */
+ ret = btrfs_commit_transaction(trans, fs_info->tree_root);
+ if (ret)
+ return ret;
+
free_extent_buffer(log_root_tree->node);
log_root_tree->log_root = NULL;
fs_info->log_root_recovering = 0;
-
- /* step 4: commit the transaction, which also unpins the blocks */
- btrfs_commit_transaction(trans, fs_info->tree_root);
-
kfree(log_root_tree);
- return 0;
+ return 0;
error:
+ if (wc.trans)
+ btrfs_end_transaction(wc.trans, fs_info->tree_root);
btrfs_free_path(path);
return ret;
}
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index 862ac813f6b..1d4ae0d15a7 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -40,9 +40,6 @@ int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
struct inode *inode, u64 dirid);
void btrfs_end_log_trans(struct btrfs_root *root);
int btrfs_pin_log_trans(struct btrfs_root *root);
-int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
- struct dentry *parent, int exists_only);
void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
struct inode *dir, struct inode *inode,
int for_rename);
diff --git a/fs/btrfs/ulist.c b/fs/btrfs/ulist.c
index 99be4c138db..7b417e20efe 100644
--- a/fs/btrfs/ulist.c
+++ b/fs/btrfs/ulist.c
@@ -5,7 +5,7 @@
*/
#include <linux/slab.h>
-#include <linux/module.h>
+#include <linux/export.h>
#include "ulist.h"
/*
@@ -53,6 +53,7 @@ void ulist_init(struct ulist *ulist)
ulist->nnodes = 0;
ulist->nodes = ulist->int_nodes;
ulist->nodes_alloced = ULIST_SIZE;
+ ulist->root = RB_ROOT;
}
EXPORT_SYMBOL(ulist_init);
@@ -72,6 +73,7 @@ void ulist_fini(struct ulist *ulist)
if (ulist->nodes_alloced > ULIST_SIZE)
kfree(ulist->nodes);
ulist->nodes_alloced = 0; /* in case ulist_fini is called twice */
+ ulist->root = RB_ROOT;
}
EXPORT_SYMBOL(ulist_fini);
@@ -123,6 +125,45 @@ void ulist_free(struct ulist *ulist)
}
EXPORT_SYMBOL(ulist_free);
+static struct ulist_node *ulist_rbtree_search(struct ulist *ulist, u64 val)
+{
+ struct rb_node *n = ulist->root.rb_node;
+ struct ulist_node *u = NULL;
+
+ while (n) {
+ u = rb_entry(n, struct ulist_node, rb_node);
+ if (u->val < val)
+ n = n->rb_right;
+ else if (u->val > val)
+ n = n->rb_left;
+ else
+ return u;
+ }
+ return NULL;
+}
+
+static int ulist_rbtree_insert(struct ulist *ulist, struct ulist_node *ins)
+{
+ struct rb_node **p = &ulist->root.rb_node;
+ struct rb_node *parent = NULL;
+ struct ulist_node *cur = NULL;
+
+ while (*p) {
+ parent = *p;
+ cur = rb_entry(parent, struct ulist_node, rb_node);
+
+ if (cur->val < ins->val)
+ p = &(*p)->rb_right;
+ else if (cur->val > ins->val)
+ p = &(*p)->rb_left;
+ else
+ return -EEXIST;
+ }
+ rb_link_node(&ins->rb_node, parent, p);
+ rb_insert_color(&ins->rb_node, &ulist->root);
+ return 0;
+}
+
/**
* ulist_add - add an element to the ulist
* @ulist: ulist to add the element to
@@ -151,14 +192,13 @@ int ulist_add(struct ulist *ulist, u64 val, u64 aux, gfp_t gfp_mask)
int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux,
u64 *old_aux, gfp_t gfp_mask)
{
- int i;
-
- for (i = 0; i < ulist->nnodes; ++i) {
- if (ulist->nodes[i].val == val) {
- if (old_aux)
- *old_aux = ulist->nodes[i].aux;
- return 0;
- }
+ int ret = 0;
+ struct ulist_node *node = NULL;
+ node = ulist_rbtree_search(ulist, val);
+ if (node) {
+ if (old_aux)
+ *old_aux = node->aux;
+ return 0;
}
if (ulist->nnodes >= ulist->nodes_alloced) {
@@ -187,6 +227,8 @@ int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux,
}
ulist->nodes[ulist->nnodes].val = val;
ulist->nodes[ulist->nnodes].aux = aux;
+ ret = ulist_rbtree_insert(ulist, &ulist->nodes[ulist->nnodes]);
+ BUG_ON(ret);
++ulist->nnodes;
return 1;
diff --git a/fs/btrfs/ulist.h b/fs/btrfs/ulist.h
index 21a1963439c..fb36731074b 100644
--- a/fs/btrfs/ulist.h
+++ b/fs/btrfs/ulist.h
@@ -8,6 +8,9 @@
#ifndef __ULIST__
#define __ULIST__
+#include <linux/list.h>
+#include <linux/rbtree.h>
+
/*
* ulist is a generic data structure to hold a collection of unique u64
* values. The only operations it supports is adding to the list and
@@ -34,6 +37,7 @@ struct ulist_iterator {
struct ulist_node {
u64 val; /* value to store */
u64 aux; /* auxiliary value saved along with the val */
+ struct rb_node rb_node; /* used to speed up search */
};
struct ulist {
@@ -54,6 +58,8 @@ struct ulist {
*/
struct ulist_node *nodes;
+ struct rb_root root;
+
/*
* inline storage space for the first ULIST_SIZE entries
*/
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 5cbb7f4b167..0e925ced971 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -25,6 +25,8 @@
#include <linux/capability.h>
#include <linux/ratelimit.h>
#include <linux/kthread.h>
+#include <linux/raid/pq.h>
+#include <asm/div64.h>
#include "compat.h"
#include "ctree.h"
#include "extent_map.h"
@@ -32,6 +34,7 @@
#include "transaction.h"
#include "print-tree.h"
#include "volumes.h"
+#include "raid56.h"
#include "async-thread.h"
#include "check-integrity.h"
#include "rcu-string.h"
@@ -43,6 +46,7 @@ static int init_first_rw_device(struct btrfs_trans_handle *trans,
struct btrfs_device *device);
static int btrfs_relocate_sys_chunks(struct btrfs_root *root);
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 DEFINE_MUTEX(uuid_mutex);
@@ -647,6 +651,7 @@ static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
new_device->writeable = 0;
new_device->in_fs_metadata = 0;
new_device->can_discard = 0;
+ spin_lock_init(&new_device->io_lock);
list_replace_rcu(&device->dev_list, &new_device->dev_list);
call_rcu(&device->rcu, free_device);
@@ -680,6 +685,12 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
__btrfs_close_devices(fs_devices);
free_fs_devices(fs_devices);
}
+ /*
+ * Wait for rcu kworkers under __btrfs_close_devices
+ * to finish all blkdev_puts so device is really
+ * free when umount is done.
+ */
+ rcu_barrier();
return ret;
}
@@ -707,9 +718,9 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
if (!device->name)
continue;
- ret = btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
- &bdev, &bh);
- if (ret)
+ /* Just open everything we can; ignore failures here */
+ if (btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
+ &bdev, &bh))
continue;
disk_super = (struct btrfs_super_block *)bh->b_data;
@@ -792,26 +803,75 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
return ret;
}
+/*
+ * Look for a btrfs signature on a device. This may be called out of the mount path
+ * and we are not allowed to call set_blocksize during the scan. The superblock
+ * is read via pagecache
+ */
int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
struct btrfs_fs_devices **fs_devices_ret)
{
struct btrfs_super_block *disk_super;
struct block_device *bdev;
- struct buffer_head *bh;
- int ret;
+ struct page *page;
+ void *p;
+ int ret = -EINVAL;
u64 devid;
u64 transid;
u64 total_devices;
+ u64 bytenr;
+ pgoff_t index;
+ /*
+ * we would like to check all the supers, but that would make
+ * a btrfs mount succeed after a mkfs from a different FS.
+ * So, we need to add a special mount option to scan for
+ * later supers, using BTRFS_SUPER_MIRROR_MAX instead
+ */
+ bytenr = btrfs_sb_offset(0);
flags |= FMODE_EXCL;
mutex_lock(&uuid_mutex);
- ret = btrfs_get_bdev_and_sb(path, flags, holder, 0, &bdev, &bh);
- if (ret)
+
+ bdev = blkdev_get_by_path(path, flags, holder);
+
+ if (IS_ERR(bdev)) {
+ ret = PTR_ERR(bdev);
goto error;
- disk_super = (struct btrfs_super_block *)bh->b_data;
+ }
+
+ /* make sure our super fits in the device */
+ if (bytenr + PAGE_CACHE_SIZE >= i_size_read(bdev->bd_inode))
+ goto error_bdev_put;
+
+ /* make sure our super fits in the page */
+ if (sizeof(*disk_super) > PAGE_CACHE_SIZE)
+ goto error_bdev_put;
+
+ /* make sure our super doesn't straddle pages on disk */
+ index = bytenr >> PAGE_CACHE_SHIFT;
+ if ((bytenr + sizeof(*disk_super) - 1) >> PAGE_CACHE_SHIFT != index)
+ goto error_bdev_put;
+
+ /* pull in the page with our super */
+ page = read_cache_page_gfp(bdev->bd_inode->i_mapping,
+ index, GFP_NOFS);
+
+ if (IS_ERR_OR_NULL(page))
+ goto error_bdev_put;
+
+ p = kmap(page);
+
+ /* align our pointer to the offset of the super block */
+ disk_super = p + (bytenr & ~PAGE_CACHE_MASK);
+
+ if (btrfs_super_bytenr(disk_super) != bytenr ||
+ disk_super->magic != cpu_to_le64(BTRFS_MAGIC))
+ goto error_unmap;
+
devid = btrfs_stack_device_id(&disk_super->dev_item);
transid = btrfs_super_generation(disk_super);
total_devices = btrfs_super_num_devices(disk_super);
+
if (disk_super->label[0]) {
if (disk_super->label[BTRFS_LABEL_SIZE - 1])
disk_super->label[BTRFS_LABEL_SIZE - 1] = '\0';
@@ -819,12 +879,19 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
} else {
printk(KERN_INFO "device fsid %pU ", disk_super->fsid);
}
+
printk(KERN_CONT "devid %llu transid %llu %s\n",
(unsigned long long)devid, (unsigned long long)transid, path);
+
ret = device_list_add(path, disk_super, devid, fs_devices_ret);
if (!ret && fs_devices_ret)
(*fs_devices_ret)->total_devices = total_devices;
- brelse(bh);
+
+error_unmap:
+ kunmap(page);
+ page_cache_release(page);
+
+error_bdev_put:
blkdev_put(bdev, flags);
error:
mutex_unlock(&uuid_mutex);
@@ -1133,10 +1200,10 @@ out:
return ret;
}
-int btrfs_alloc_dev_extent(struct btrfs_trans_handle *trans,
- struct btrfs_device *device,
- u64 chunk_tree, u64 chunk_objectid,
- u64 chunk_offset, u64 start, u64 num_bytes)
+static int btrfs_alloc_dev_extent(struct btrfs_trans_handle *trans,
+ struct btrfs_device *device,
+ u64 chunk_tree, u64 chunk_objectid,
+ u64 chunk_offset, u64 start, u64 num_bytes)
{
int ret;
struct btrfs_path *path;
@@ -1263,9 +1330,9 @@ error:
* the device information is stored in the chunk root
* the btrfs_device struct should be fully filled in
*/
-int btrfs_add_device(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_device *device)
+static int btrfs_add_device(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct btrfs_device *device)
{
int ret;
struct btrfs_path *path;
@@ -1372,14 +1439,19 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
u64 devid;
u64 num_devices;
u8 *dev_uuid;
+ unsigned seq;
int ret = 0;
bool clear_super = false;
mutex_lock(&uuid_mutex);
- all_avail = root->fs_info->avail_data_alloc_bits |
- root->fs_info->avail_system_alloc_bits |
- root->fs_info->avail_metadata_alloc_bits;
+ do {
+ seq = read_seqbegin(&root->fs_info->profiles_lock);
+
+ all_avail = root->fs_info->avail_data_alloc_bits |
+ root->fs_info->avail_system_alloc_bits |
+ root->fs_info->avail_metadata_alloc_bits;
+ } while (read_seqretry(&root->fs_info->profiles_lock, seq));
num_devices = root->fs_info->fs_devices->num_devices;
btrfs_dev_replace_lock(&root->fs_info->dev_replace);
@@ -1403,6 +1475,21 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
goto out;
}
+ if ((all_avail & BTRFS_BLOCK_GROUP_RAID5) &&
+ root->fs_info->fs_devices->rw_devices <= 2) {
+ printk(KERN_ERR "btrfs: unable to go below two "
+ "devices on raid5\n");
+ ret = -EINVAL;
+ goto out;
+ }
+ if ((all_avail & BTRFS_BLOCK_GROUP_RAID6) &&
+ root->fs_info->fs_devices->rw_devices <= 3) {
+ printk(KERN_ERR "btrfs: unable to go below three "
+ "devices on raid6\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
if (strcmp(device_path, "missing") == 0) {
struct list_head *devices;
struct btrfs_device *tmp;
@@ -1624,8 +1711,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
}
-int btrfs_find_device_by_path(struct btrfs_root *root, char *device_path,
- struct btrfs_device **device)
+static int btrfs_find_device_by_path(struct btrfs_root *root, char *device_path,
+ struct btrfs_device **device)
{
int ret = 0;
struct btrfs_super_block *disk_super;
@@ -2299,7 +2386,11 @@ static int btrfs_relocate_chunk(struct btrfs_root *root,
return ret;
trans = btrfs_start_transaction(root, 0);
- BUG_ON(IS_ERR(trans));
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ btrfs_std_error(root->fs_info, ret);
+ return ret;
+ }
lock_chunks(root);
@@ -2616,7 +2707,7 @@ static int chunk_usage_filter(struct btrfs_fs_info *fs_info, u64 chunk_offset,
chunk_used = btrfs_block_group_used(&cache->item);
if (bargs->usage == 0)
- user_thresh = 0;
+ user_thresh = 1;
else if (bargs->usage > 100)
user_thresh = cache->key.offset;
else
@@ -2664,11 +2755,15 @@ static int chunk_drange_filter(struct extent_buffer *leaf,
return 0;
if (btrfs_chunk_type(leaf, chunk) & (BTRFS_BLOCK_GROUP_DUP |
- BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10))
- factor = 2;
- else
- factor = 1;
- factor = num_stripes / factor;
+ BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10)) {
+ factor = num_stripes / 2;
+ } else if (btrfs_chunk_type(leaf, chunk) & BTRFS_BLOCK_GROUP_RAID5) {
+ factor = num_stripes - 1;
+ } else if (btrfs_chunk_type(leaf, chunk) & BTRFS_BLOCK_GROUP_RAID6) {
+ factor = num_stripes - 2;
+ } else {
+ factor = num_stripes;
+ }
for (i = 0; i < num_stripes; i++) {
stripe = btrfs_stripe_nr(chunk, i);
@@ -2966,7 +3061,8 @@ static void __cancel_balance(struct btrfs_fs_info *fs_info)
unset_balance_control(fs_info);
ret = del_balance_item(fs_info->tree_root);
- BUG_ON(ret);
+ if (ret)
+ btrfs_std_error(fs_info, ret);
atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
}
@@ -2985,6 +3081,7 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
int mixed = 0;
int ret;
u64 num_devices;
+ unsigned seq;
if (btrfs_fs_closing(fs_info) ||
atomic_read(&fs_info->balance_pause_req) ||
@@ -3027,7 +3124,9 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
allowed |= (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID1);
else
allowed |= (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID1 |
- BTRFS_BLOCK_GROUP_RAID10);
+ BTRFS_BLOCK_GROUP_RAID10 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6);
if ((bctl->data.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
(!alloc_profile_is_valid(bctl->data.target, 1) ||
@@ -3067,23 +3166,29 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
/* allow to reduce meta or sys integrity only if force set */
allowed = BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
- BTRFS_BLOCK_GROUP_RAID10;
- if (((bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
- (fs_info->avail_system_alloc_bits & allowed) &&
- !(bctl->sys.target & allowed)) ||
- ((bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
- (fs_info->avail_metadata_alloc_bits & allowed) &&
- !(bctl->meta.target & allowed))) {
- if (bctl->flags & BTRFS_BALANCE_FORCE) {
- printk(KERN_INFO "btrfs: force reducing metadata "
- "integrity\n");
- } else {
- printk(KERN_ERR "btrfs: balance will reduce metadata "
- "integrity, use force if you want this\n");
- ret = -EINVAL;
- goto out;
+ BTRFS_BLOCK_GROUP_RAID10 |
+ BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6;
+ do {
+ seq = read_seqbegin(&fs_info->profiles_lock);
+
+ if (((bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
+ (fs_info->avail_system_alloc_bits & allowed) &&
+ !(bctl->sys.target & allowed)) ||
+ ((bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
+ (fs_info->avail_metadata_alloc_bits & allowed) &&
+ !(bctl->meta.target & allowed))) {
+ if (bctl->flags & BTRFS_BALANCE_FORCE) {
+ printk(KERN_INFO "btrfs: force reducing metadata "
+ "integrity\n");
+ } else {
+ printk(KERN_ERR "btrfs: balance will reduce metadata "
+ "integrity, use force if you want this\n");
+ ret = -EINVAL;
+ goto out;
+ }
}
- }
+ } while (read_seqretry(&fs_info->profiles_lock, seq));
if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
int num_tolerated_disk_barrier_failures;
@@ -3127,6 +3232,11 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
mutex_lock(&fs_info->balance_mutex);
atomic_dec(&fs_info->balance_running);
+ if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
+ fs_info->num_tolerated_disk_barrier_failures =
+ btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
+ }
+
if (bargs) {
memset(bargs, 0, sizeof(*bargs));
update_ioctl_balance_args(fs_info, 0, bargs);
@@ -3137,11 +3247,6 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
__cancel_balance(fs_info);
}
- if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
- fs_info->num_tolerated_disk_barrier_failures =
- btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
- }
-
wake_up(&fs_info->balance_wait_q);
return ret;
@@ -3503,14 +3608,79 @@ static int btrfs_cmp_device_info(const void *a, const void *b)
return 0;
}
-struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
- { 2, 1, 0, 4, 2, 2 /* raid10 */ },
- { 1, 1, 2, 2, 2, 2 /* raid1 */ },
- { 1, 2, 1, 1, 1, 2 /* dup */ },
- { 1, 1, 0, 2, 1, 1 /* raid0 */ },
- { 1, 1, 1, 1, 1, 1 /* single */ },
+static struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
+ [BTRFS_RAID_RAID10] = {
+ .sub_stripes = 2,
+ .dev_stripes = 1,
+ .devs_max = 0, /* 0 == as many as possible */
+ .devs_min = 4,
+ .devs_increment = 2,
+ .ncopies = 2,
+ },
+ [BTRFS_RAID_RAID1] = {
+ .sub_stripes = 1,
+ .dev_stripes = 1,
+ .devs_max = 2,
+ .devs_min = 2,
+ .devs_increment = 2,
+ .ncopies = 2,
+ },
+ [BTRFS_RAID_DUP] = {
+ .sub_stripes = 1,
+ .dev_stripes = 2,
+ .devs_max = 1,
+ .devs_min = 1,
+ .devs_increment = 1,
+ .ncopies = 2,
+ },
+ [BTRFS_RAID_RAID0] = {
+ .sub_stripes = 1,
+ .dev_stripes = 1,
+ .devs_max = 0,
+ .devs_min = 2,
+ .devs_increment = 1,
+ .ncopies = 1,
+ },
+ [BTRFS_RAID_SINGLE] = {
+ .sub_stripes = 1,
+ .dev_stripes = 1,
+ .devs_max = 1,
+ .devs_min = 1,
+ .devs_increment = 1,
+ .ncopies = 1,
+ },
+ [BTRFS_RAID_RAID5] = {
+ .sub_stripes = 1,
+ .dev_stripes = 1,
+ .devs_max = 0,
+ .devs_min = 2,
+ .devs_increment = 1,
+ .ncopies = 2,
+ },
+ [BTRFS_RAID_RAID6] = {
+ .sub_stripes = 1,
+ .dev_stripes = 1,
+ .devs_max = 0,
+ .devs_min = 3,
+ .devs_increment = 1,
+ .ncopies = 3,
+ },
};
+static u32 find_raid56_stripe_len(u32 data_devices, u32 dev_stripe_target)
+{
+ /* TODO allow them to set a preferred stripe size */
+ return 64 * 1024;
+}
+
+static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
+{
+ if (!(type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)))
+ return;
+
+ btrfs_set_fs_incompat(info, RAID56);
+}
+
static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
struct btrfs_root *extent_root,
struct map_lookup **map_ret,
@@ -3526,6 +3696,8 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
struct btrfs_device_info *devices_info = NULL;
u64 total_avail;
int num_stripes; /* total number of stripes to allocate */
+ int data_stripes; /* number of stripes that count for
+ block group size */
int sub_stripes; /* sub_stripes info for map */
int dev_stripes; /* stripes per dev */
int devs_max; /* max devs to use */
@@ -3537,6 +3709,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
u64 max_chunk_size;
u64 stripe_size;
u64 num_bytes;
+ u64 raid_stripe_len = BTRFS_STRIPE_LEN;
int ndevs;
int i;
int j;
@@ -3631,12 +3804,16 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
if (max_avail < BTRFS_STRIPE_LEN * dev_stripes)
continue;
+ if (ndevs == fs_devices->rw_devices) {
+ WARN(1, "%s: found more than %llu devices\n",
+ __func__, fs_devices->rw_devices);
+ break;
+ }
devices_info[ndevs].dev_offset = dev_offset;
devices_info[ndevs].max_avail = max_avail;
devices_info[ndevs].total_avail = total_avail;
devices_info[ndevs].dev = device;
++ndevs;
- WARN_ON(ndevs > fs_devices->rw_devices);
}
/*
@@ -3662,16 +3839,48 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
stripe_size = devices_info[ndevs-1].max_avail;
num_stripes = ndevs * dev_stripes;
- if (stripe_size * ndevs > max_chunk_size * ncopies) {
- stripe_size = max_chunk_size * ncopies;
- do_div(stripe_size, ndevs);
+ /*
+ * this will have to be fixed for RAID1 and RAID10 over
+ * more drives
+ */
+ data_stripes = num_stripes / ncopies;
+
+ if (type & BTRFS_BLOCK_GROUP_RAID5) {
+ raid_stripe_len = find_raid56_stripe_len(ndevs - 1,
+ btrfs_super_stripesize(info->super_copy));
+ data_stripes = num_stripes - 1;
+ }
+ if (type & BTRFS_BLOCK_GROUP_RAID6) {
+ raid_stripe_len = find_raid56_stripe_len(ndevs - 2,
+ btrfs_super_stripesize(info->super_copy));
+ data_stripes = num_stripes - 2;
+ }
+
+ /*
+ * Use the number of data stripes to figure out how big this chunk
+ * is really going to be in terms of logical address space,
+ * and compare that answer with the max chunk size
+ */
+ if (stripe_size * data_stripes > max_chunk_size) {
+ u64 mask = (1ULL << 24) - 1;
+ stripe_size = max_chunk_size;
+ do_div(stripe_size, data_stripes);
+
+ /* bump the answer up to a 16MB boundary */
+ stripe_size = (stripe_size + mask) & ~mask;
+
+ /* but don't go higher than the limits we found
+ * while searching for free extents
+ */
+ if (stripe_size > devices_info[ndevs-1].max_avail)
+ stripe_size = devices_info[ndevs-1].max_avail;
}
do_div(stripe_size, dev_stripes);
/* align to BTRFS_STRIPE_LEN */
- do_div(stripe_size, BTRFS_STRIPE_LEN);
- stripe_size *= BTRFS_STRIPE_LEN;
+ do_div(stripe_size, raid_stripe_len);
+ stripe_size *= raid_stripe_len;
map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
if (!map) {
@@ -3689,14 +3898,14 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
}
}
map->sector_size = extent_root->sectorsize;
- map->stripe_len = BTRFS_STRIPE_LEN;
- map->io_align = BTRFS_STRIPE_LEN;
- map->io_width = BTRFS_STRIPE_LEN;
+ map->stripe_len = raid_stripe_len;
+ map->io_align = raid_stripe_len;
+ map->io_width = raid_stripe_len;
map->type = type;
map->sub_stripes = sub_stripes;
*map_ret = map;
- num_bytes = stripe_size * (num_stripes / ncopies);
+ num_bytes = stripe_size * data_stripes;
*stripe_size_out = stripe_size;
*num_bytes_out = num_bytes;
@@ -3716,17 +3925,12 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
em_tree = &extent_root->fs_info->mapping_tree.map_tree;
write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
+ ret = add_extent_mapping(em_tree, em, 0);
write_unlock(&em_tree->lock);
- free_extent_map(em);
- if (ret)
- goto error;
-
- ret = btrfs_make_block_group(trans, extent_root, 0, type,
- BTRFS_FIRST_CHUNK_TREE_OBJECTID,
- start, num_bytes);
- if (ret)
+ if (ret) {
+ free_extent_map(em);
goto error;
+ }
for (i = 0; i < map->num_stripes; ++i) {
struct btrfs_device *device;
@@ -3739,15 +3943,44 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
info->chunk_root->root_key.objectid,
BTRFS_FIRST_CHUNK_TREE_OBJECTID,
start, dev_offset, stripe_size);
- if (ret) {
- btrfs_abort_transaction(trans, extent_root, ret);
- goto error;
- }
+ if (ret)
+ goto error_dev_extent;
+ }
+
+ ret = btrfs_make_block_group(trans, extent_root, 0, type,
+ BTRFS_FIRST_CHUNK_TREE_OBJECTID,
+ start, num_bytes);
+ if (ret) {
+ i = map->num_stripes - 1;
+ goto error_dev_extent;
}
+ free_extent_map(em);
+ check_raid56_incompat_flag(extent_root->fs_info, type);
+
kfree(devices_info);
return 0;
+error_dev_extent:
+ for (; i >= 0; i--) {
+ struct btrfs_device *device;
+ int err;
+
+ device = map->stripes[i].dev;
+ err = btrfs_free_dev_extent(trans, device, start);
+ if (err) {
+ btrfs_abort_transaction(trans, extent_root, err);
+ break;
+ }
+ }
+ write_lock(&em_tree->lock);
+ remove_extent_mapping(em_tree, em);
+ write_unlock(&em_tree->lock);
+
+ /* One for our allocation */
+ free_extent_map(em);
+ /* One for the tree reference */
+ free_extent_map(em);
error:
kfree(map);
kfree(devices_info);
@@ -3887,10 +4120,7 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans,
if (ret)
return ret;
- alloc_profile = BTRFS_BLOCK_GROUP_METADATA |
- fs_info->avail_metadata_alloc_bits;
- alloc_profile = btrfs_reduce_alloc_profile(root, alloc_profile);
-
+ alloc_profile = btrfs_get_alloc_profile(extent_root, 0);
ret = __btrfs_alloc_chunk(trans, extent_root, &map, &chunk_size,
&stripe_size, chunk_offset, alloc_profile);
if (ret)
@@ -3898,10 +4128,7 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans,
sys_chunk_offset = chunk_offset + chunk_size;
- alloc_profile = BTRFS_BLOCK_GROUP_SYSTEM |
- fs_info->avail_system_alloc_bits;
- alloc_profile = btrfs_reduce_alloc_profile(root, alloc_profile);
-
+ alloc_profile = btrfs_get_alloc_profile(fs_info->chunk_root, 0);
ret = __btrfs_alloc_chunk(trans, extent_root, &sys_map,
&sys_chunk_size, &sys_stripe_size,
sys_chunk_offset, alloc_profile);
@@ -4006,14 +4233,34 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
read_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, logical, len);
read_unlock(&em_tree->lock);
- BUG_ON(!em);
- BUG_ON(em->start > logical || em->start + em->len < logical);
+ /*
+ * 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_emerg(fs_info, "No mapping for %Lu-%Lu\n", logical,
+ logical+len);
+ return 1;
+ }
+
+ if (em->start > logical || em->start + em->len < logical) {
+ btrfs_emerg(fs_info, "Invalid mapping for %Lu-%Lu, got "
+ "%Lu-%Lu\n", logical, logical+len, em->start,
+ em->start + em->len);
+ return 1;
+ }
+
map = (struct map_lookup *)em->bdev;
if (map->type & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1))
ret = map->num_stripes;
else if (map->type & BTRFS_BLOCK_GROUP_RAID10)
ret = map->sub_stripes;
+ else if (map->type & BTRFS_BLOCK_GROUP_RAID5)
+ ret = 2;
+ else if (map->type & BTRFS_BLOCK_GROUP_RAID6)
+ ret = 3;
else
ret = 1;
free_extent_map(em);
@@ -4026,6 +4273,52 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
return ret;
}
+unsigned long btrfs_full_stripe_len(struct btrfs_root *root,
+ struct btrfs_mapping_tree *map_tree,
+ u64 logical)
+{
+ struct extent_map *em;
+ struct map_lookup *map;
+ struct extent_map_tree *em_tree = &map_tree->map_tree;
+ unsigned long len = root->sectorsize;
+
+ read_lock(&em_tree->lock);
+ em = lookup_extent_mapping(em_tree, logical, len);
+ read_unlock(&em_tree->lock);
+ BUG_ON(!em);
+
+ BUG_ON(em->start > logical || em->start + em->len < logical);
+ map = (struct map_lookup *)em->bdev;
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ len = map->stripe_len * nr_data_stripes(map);
+ }
+ free_extent_map(em);
+ return len;
+}
+
+int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
+ 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);
+
+ BUG_ON(em->start > logical || em->start + em->len < logical);
+ map = (struct map_lookup *)em->bdev;
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6))
+ ret = 1;
+ free_extent_map(em);
+ return ret;
+}
+
static int find_live_mirror(struct btrfs_fs_info *fs_info,
struct map_lookup *map, int first, int num,
int optimal, int dev_replace_is_ongoing)
@@ -4063,10 +4356,39 @@ static int find_live_mirror(struct btrfs_fs_info *fs_info,
return optimal;
}
+static inline int parity_smaller(u64 a, u64 b)
+{
+ return a > b;
+}
+
+/* Bubble-sort the stripe set to put the parity/syndrome stripes last */
+static void sort_parity_stripes(struct btrfs_bio *bbio, u64 *raid_map)
+{
+ struct btrfs_bio_stripe s;
+ int i;
+ u64 l;
+ int again = 1;
+
+ while (again) {
+ again = 0;
+ for (i = 0; i < bbio->num_stripes - 1; i++) {
+ if (parity_smaller(raid_map[i], raid_map[i+1])) {
+ s = bbio->stripes[i];
+ l = raid_map[i];
+ bbio->stripes[i] = bbio->stripes[i+1];
+ raid_map[i] = raid_map[i+1];
+ bbio->stripes[i+1] = s;
+ raid_map[i+1] = l;
+ again = 1;
+ }
+ }
+ }
+}
+
static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
u64 logical, u64 *length,
struct btrfs_bio **bbio_ret,
- int mirror_num)
+ int mirror_num, u64 **raid_map_ret)
{
struct extent_map *em;
struct map_lookup *map;
@@ -4078,6 +4400,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
u64 stripe_nr;
u64 stripe_nr_orig;
u64 stripe_nr_end;
+ u64 stripe_len;
+ u64 *raid_map = NULL;
int stripe_index;
int i;
int ret = 0;
@@ -4089,45 +4413,86 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
int num_alloc_stripes;
int patch_the_first_stripe_for_dev_replace = 0;
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) {
- printk(KERN_CRIT "btrfs: unable to find logical %llu len %llu\n",
- (unsigned long long)logical,
- (unsigned long long)*length);
- BUG();
+ btrfs_crit(fs_info, "unable to find logical %llu len %llu",
+ (unsigned long long)logical,
+ (unsigned long long)*length);
+ return -EINVAL;
+ }
+
+ if (em->start > logical || em->start + em->len < logical) {
+ btrfs_crit(fs_info, "found a bad mapping, wanted %Lu, "
+ "found %Lu-%Lu\n", logical, em->start,
+ em->start + em->len);
+ return -EINVAL;
}
- BUG_ON(em->start > logical || em->start + em->len < logical);
map = (struct map_lookup *)em->bdev;
offset = logical - em->start;
+ if (mirror_num > map->num_stripes)
+ mirror_num = 0;
+
+ stripe_len = map->stripe_len;
stripe_nr = offset;
/*
* stripe_nr counts the total number of stripes we have to stride
* to get to this block
*/
- do_div(stripe_nr, map->stripe_len);
+ do_div(stripe_nr, stripe_len);
- stripe_offset = stripe_nr * map->stripe_len;
+ stripe_offset = stripe_nr * stripe_len;
BUG_ON(offset < stripe_offset);
/* stripe_offset is the offset of this block in its stripe*/
stripe_offset = offset - stripe_offset;
- if (rw & REQ_DISCARD)
+ /* if we're here for raid56, we need to know the stripe aligned start */
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)) {
+ unsigned long full_stripe_len = stripe_len * nr_data_stripes(map);
+ raid56_full_stripe_start = offset;
+
+ /* allow a write of a full stripe, but make sure we don't
+ * allow straddling of stripes
+ */
+ do_div(raid56_full_stripe_start, full_stripe_len);
+ raid56_full_stripe_start *= full_stripe_len;
+ }
+
+ if (rw & REQ_DISCARD) {
+ /* we don't discard raid56 yet */
+ if (map->type &
+ (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)) {
+ ret = -EOPNOTSUPP;
+ goto out;
+ }
*length = min_t(u64, em->len - offset, *length);
- else if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
- /* we limit the length of each bio to what fits in a stripe */
- *length = min_t(u64, em->len - offset,
- map->stripe_len - stripe_offset);
+ } else 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
+ stripe (on a single disk). */
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6) &&
+ (rw & REQ_WRITE)) {
+ max_len = stripe_len * nr_data_stripes(map) -
+ (offset - raid56_full_stripe_start);
+ } else {
+ /* we limit the length of each bio to what fits in a stripe */
+ max_len = stripe_len - stripe_offset;
+ }
+ *length = min_t(u64, em->len - offset, max_len);
} else {
*length = em->len - offset;
}
+ /* This is for when we're called from btrfs_merge_bio_hook() and all
+ it cares about is the length */
if (!bbio_ret)
goto out;
@@ -4160,7 +4525,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
u64 physical_of_found = 0;
ret = __btrfs_map_block(fs_info, REQ_GET_READ_MIRRORS,
- logical, &tmp_length, &tmp_bbio, 0);
+ logical, &tmp_length, &tmp_bbio, 0, NULL);
if (ret) {
WARN_ON(tmp_bbio != NULL);
goto out;
@@ -4221,11 +4586,11 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
num_stripes = 1;
stripe_index = 0;
stripe_nr_orig = stripe_nr;
- stripe_nr_end = (offset + *length + map->stripe_len - 1) &
- (~(map->stripe_len - 1));
+ stripe_nr_end = ALIGN(offset + *length, map->stripe_len);
do_div(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 (rw & REQ_DISCARD)
num_stripes = min_t(u64, map->num_stripes,
@@ -4276,6 +4641,65 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
dev_replace_is_ongoing);
mirror_num = stripe_index - old_stripe_index + 1;
}
+
+ } else if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ u64 tmp;
+
+ if (bbio_ret && ((rw & REQ_WRITE) || mirror_num > 1)
+ && raid_map_ret) {
+ int i, rot;
+
+ /* push stripe_nr back to the start of the full stripe */
+ stripe_nr = raid56_full_stripe_start;
+ do_div(stripe_nr, stripe_len);
+
+ stripe_index = do_div(stripe_nr, nr_data_stripes(map));
+
+ /* RAID[56] write or recovery. Return all stripes */
+ num_stripes = map->num_stripes;
+ max_errors = nr_parity_stripes(map);
+
+ raid_map = kmalloc(sizeof(u64) * num_stripes,
+ GFP_NOFS);
+ if (!raid_map) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ /* Work out the disk rotation on this stripe-set */
+ tmp = stripe_nr;
+ rot = do_div(tmp, num_stripes);
+
+ /* Fill in the logical address of each stripe */
+ tmp = stripe_nr * nr_data_stripes(map);
+ for (i = 0; i < nr_data_stripes(map); i++)
+ raid_map[(i+rot) % num_stripes] =
+ em->start + (tmp + i) * map->stripe_len;
+
+ raid_map[(i+rot) % map->num_stripes] = RAID5_P_STRIPE;
+ if (map->type & BTRFS_BLOCK_GROUP_RAID6)
+ raid_map[(i+rot+1) % num_stripes] =
+ RAID6_Q_STRIPE;
+
+ *length = map->stripe_len;
+ stripe_index = 0;
+ stripe_offset = 0;
+ } else {
+ /*
+ * Mirror #0 or #1 means the original data block.
+ * Mirror #2 is RAID5 parity block.
+ * Mirror #3 is RAID6 Q block.
+ */
+ stripe_index = do_div(stripe_nr, nr_data_stripes(map));
+ if (mirror_num > 1)
+ stripe_index = nr_data_stripes(map) +
+ mirror_num - 2;
+
+ /* We distribute the parity blocks across stripes */
+ tmp = stripe_nr + stripe_index;
+ stripe_index = do_div(tmp, map->num_stripes);
+ }
} else {
/*
* after this do_div call, stripe_nr is the number of stripes
@@ -4384,8 +4808,11 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
if (rw & (REQ_WRITE | REQ_GET_READ_MIRRORS)) {
if (map->type & (BTRFS_BLOCK_GROUP_RAID1 |
BTRFS_BLOCK_GROUP_RAID10 |
+ BTRFS_BLOCK_GROUP_RAID5 |
BTRFS_BLOCK_GROUP_DUP)) {
max_errors = 1;
+ } else if (map->type & BTRFS_BLOCK_GROUP_RAID6) {
+ max_errors = 2;
}
}
@@ -4486,6 +4913,10 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
bbio->stripes[0].physical = physical_to_patch_in_first_stripe;
bbio->mirror_num = map->num_stripes + 1;
}
+ if (raid_map) {
+ sort_parity_stripes(bbio, raid_map);
+ *raid_map_ret = raid_map;
+ }
out:
if (dev_replace_is_ongoing)
btrfs_dev_replace_unlock(dev_replace);
@@ -4498,7 +4929,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
struct btrfs_bio **bbio_ret, int mirror_num)
{
return __btrfs_map_block(fs_info, rw, logical, length, bbio_ret,
- mirror_num);
+ mirror_num, NULL);
}
int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
@@ -4512,20 +4943,39 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
u64 bytenr;
u64 length;
u64 stripe_nr;
+ 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);
- BUG_ON(!em || em->start != chunk_start);
+ if (!em) {
+ printk(KERN_ERR "btrfs: couldn't find em for chunk %Lu\n",
+ chunk_start);
+ return -EIO;
+ }
+
+ if (em->start != chunk_start) {
+ printk(KERN_ERR "btrfs: bad chunk start, em=%Lu, wanted=%Lu\n",
+ em->start, chunk_start);
+ free_extent_map(em);
+ return -EIO;
+ }
map = (struct map_lookup *)em->bdev;
length = em->len;
+ rmap_len = map->stripe_len;
+
if (map->type & BTRFS_BLOCK_GROUP_RAID10)
do_div(length, map->num_stripes / map->sub_stripes);
else if (map->type & BTRFS_BLOCK_GROUP_RAID0)
do_div(length, map->num_stripes);
+ else if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ do_div(length, nr_data_stripes(map));
+ rmap_len = map->stripe_len * nr_data_stripes(map);
+ }
buf = kzalloc(sizeof(u64) * map->num_stripes, GFP_NOFS);
BUG_ON(!buf); /* -ENOMEM */
@@ -4545,8 +4995,11 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
do_div(stripe_nr, map->sub_stripes);
} else if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
stripe_nr = stripe_nr * map->num_stripes + i;
- }
- bytenr = chunk_start + stripe_nr * map->stripe_len;
+ } /* else if RAID[56], multiply by nr_data_stripes().
+ * Alternatively, just use rmap_len below instead of
+ * map->stripe_len */
+
+ bytenr = chunk_start + stripe_nr * rmap_len;
WARN_ON(nr >= map->num_stripes);
for (j = 0; j < nr; j++) {
if (buf[j] == bytenr)
@@ -4560,7 +5013,7 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
*logical = buf;
*naddrs = nr;
- *stripe_len = map->stripe_len;
+ *stripe_len = rmap_len;
free_extent_map(em);
return 0;
@@ -4634,7 +5087,7 @@ static void btrfs_end_bio(struct bio *bio, int err)
bio->bi_bdev = (struct block_device *)
(unsigned long)bbio->mirror_num;
/* only send an error to the higher layers if it is
- * beyond the tolerance of the multi-bio
+ * beyond the tolerance of the btrfs bio
*/
if (atomic_read(&bbio->error) > bbio->max_errors) {
err = -EIO;
@@ -4668,13 +5121,18 @@ struct async_sched {
* This will add one bio to the pending list for a device and make sure
* the work struct is scheduled.
*/
-static noinline void schedule_bio(struct btrfs_root *root,
- struct btrfs_device *device,
- int rw, struct bio *bio)
+static noinline void btrfs_schedule_bio(struct btrfs_root *root,
+ struct btrfs_device *device,
+ int rw, struct bio *bio)
{
int should_queue = 1;
struct btrfs_pending_bios *pending_bios;
+ if (device->missing || !device->bdev) {
+ bio_endio(bio, -EIO);
+ return;
+ }
+
/* don't bother with additional async steps for reads, right now */
if (!(rw & REQ_WRITE)) {
bio_get(bio);
@@ -4734,7 +5192,7 @@ static int bio_size_ok(struct block_device *bdev, struct bio *bio,
}
prev = &bio->bi_io_vec[bio->bi_vcnt - 1];
- if ((bio->bi_size >> 9) > max_sectors)
+ if (bio_sectors(bio) > max_sectors)
return 0;
if (!q->merge_bvec_fn)
@@ -4772,7 +5230,7 @@ static void submit_stripe_bio(struct btrfs_root *root, struct btrfs_bio *bbio,
#endif
bio->bi_bdev = dev->bdev;
if (async)
- schedule_bio(root, dev, rw, bio);
+ btrfs_schedule_bio(root, dev, rw, bio);
else
btrfsic_submit_bio(rw, bio);
}
@@ -4831,6 +5289,7 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
u64 logical = (u64)bio->bi_sector << 9;
u64 length = 0;
u64 map_length;
+ u64 *raid_map = NULL;
int ret;
int dev_nr = 0;
int total_devs = 1;
@@ -4839,25 +5298,38 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
length = bio->bi_size;
map_length = length;
- ret = btrfs_map_block(root->fs_info, rw, logical, &map_length, &bbio,
- mirror_num);
- if (ret)
+ ret = __btrfs_map_block(root->fs_info, rw, logical, &map_length, &bbio,
+ mirror_num, &raid_map);
+ if (ret) /* -ENOMEM */
return ret;
total_devs = bbio->num_stripes;
- if (map_length < length) {
- printk(KERN_CRIT "btrfs: mapping failed logical %llu bio len %llu "
- "len %llu\n", (unsigned long long)logical,
- (unsigned long long)length,
- (unsigned long long)map_length);
- BUG();
- }
-
bbio->orig_bio = first_bio;
bbio->private = first_bio->bi_private;
bbio->end_io = first_bio->bi_end_io;
atomic_set(&bbio->stripes_pending, bbio->num_stripes);
+ if (raid_map) {
+ /* In this case, map_length has been set to the length of
+ a single stripe; not the whole write */
+ if (rw & WRITE) {
+ return raid56_parity_write(root, bio, bbio,
+ raid_map, map_length);
+ } else {
+ return raid56_parity_recover(root, bio, bbio,
+ raid_map, map_length,
+ mirror_num);
+ }
+ }
+
+ if (map_length < length) {
+ btrfs_crit(root->fs_info, "mapping failed logical %llu bio len %llu len %llu",
+ (unsigned long long)logical,
+ (unsigned long long)length,
+ (unsigned long long)map_length);
+ BUG();
+ }
+
while (dev_nr < total_devs) {
dev = bbio->stripes[dev_nr].dev;
if (!dev || !dev->bdev || (rw & WRITE && !dev->writeable)) {
@@ -5019,7 +5491,7 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
}
write_lock(&map_tree->map_tree.lock);
- ret = add_extent_mapping(&map_tree->map_tree, em);
+ ret = add_extent_mapping(&map_tree->map_tree, em, 0);
write_unlock(&map_tree->map_tree.lock);
BUG_ON(ret); /* Tree corruption */
free_extent_map(em);
@@ -5126,8 +5598,8 @@ static int read_one_dev(struct btrfs_root *root,
return -EIO;
if (!device) {
- printk(KERN_WARNING "warning devid %llu missing\n",
- (unsigned long long)devid);
+ btrfs_warn(root->fs_info, "devid %llu missing",
+ (unsigned long long)devid);
device = add_missing_dev(root, devid, dev_uuid);
if (!device)
return -ENOMEM;
@@ -5469,7 +5941,7 @@ void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index)
btrfs_dev_stat_print_on_error(dev);
}
-void btrfs_dev_stat_print_on_error(struct btrfs_device *dev)
+static void btrfs_dev_stat_print_on_error(struct btrfs_device *dev)
{
if (!dev->dev_stats_valid)
return;
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index d3c3939ac75..845ccbb0d2e 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -21,8 +21,8 @@
#include <linux/bio.h>
#include <linux/sort.h>
+#include <linux/btrfs.h>
#include "async-thread.h"
-#include "ioctl.h"
#define BTRFS_STRIPE_LEN (64 * 1024)
@@ -254,10 +254,6 @@ int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
#define btrfs_bio_size(n) (sizeof(struct btrfs_bio) + \
(sizeof(struct btrfs_bio_stripe) * (n)))
-int btrfs_alloc_dev_extent(struct btrfs_trans_handle *trans,
- struct btrfs_device *device,
- u64 chunk_tree, u64 chunk_objectid,
- u64 chunk_offset, u64 start, u64 num_bytes);
int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
u64 logical, u64 *length,
struct btrfs_bio **bbio_ret, int mirror_num);
@@ -282,11 +278,6 @@ void btrfs_close_extra_devices(struct btrfs_fs_info *fs_info,
int btrfs_find_device_missing_or_by_path(struct btrfs_root *root,
char *device_path,
struct btrfs_device **device);
-int btrfs_find_device_by_path(struct btrfs_root *root, char *device_path,
- struct btrfs_device **device);
-int btrfs_add_device(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_device *device);
int btrfs_rm_device(struct btrfs_root *root, char *device_path);
void btrfs_cleanup_fs_uuids(void);
int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
@@ -307,7 +298,6 @@ int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
int btrfs_chunk_readonly(struct btrfs_root *root, u64 chunk_offset);
int find_free_dev_extent(struct btrfs_device *device, u64 num_bytes,
u64 *start, u64 *max_avail);
-void btrfs_dev_stat_print_on_error(struct btrfs_device *device);
void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
int btrfs_get_dev_stats(struct btrfs_root *root,
struct btrfs_ioctl_get_dev_stats *stats);
@@ -321,7 +311,11 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
struct btrfs_device *tgtdev);
int btrfs_scratch_superblock(struct btrfs_device *device);
-
+int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
+ u64 logical, u64 len, int mirror_num);
+unsigned long btrfs_full_stripe_len(struct btrfs_root *root,
+ struct btrfs_mapping_tree *map_tree,
+ u64 logical);
static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
int index)
{
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 446a6848c55..05740b9789e 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -406,8 +406,8 @@ int btrfs_removexattr(struct dentry *dentry, const char *name)
XATTR_REPLACE);
}
-int btrfs_initxattrs(struct inode *inode, const struct xattr *xattr_array,
- void *fs_info)
+static int btrfs_initxattrs(struct inode *inode,
+ const struct xattr *xattr_array, void *fs_info)
{
const struct xattr *xattr;
struct btrfs_trans_handle *trans = fs_info;