#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
-#include "volumes.h"
#include "ordered-data.h"
#include "xattr.h"
#include "tree-log.h"
+#include "volumes.h"
#include "compression.h"
#include "locking.h"
#include "free-space-cache.h"
(BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS))) {
WARN_ON(pages);
pages = kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
- BUG_ON(!pages);
+ if (!pages) {
+ /* just bail out to the uncompressed code */
+ goto cont;
+ }
if (BTRFS_I(inode)->force_compress)
compress_type = BTRFS_I(inode)->force_compress;
will_compress = 1;
}
}
+cont:
if (start == 0) {
trans = btrfs_join_transaction(root);
BUG_ON(IS_ERR(trans));
}
BUG_ON(disk_num_bytes >
- btrfs_super_total_bytes(&root->fs_info->super_copy));
+ btrfs_super_total_bytes(root->fs_info->super_copy));
alloc_hint = get_extent_allocation_hint(inode, start, num_bytes);
btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0);
}
ret = 0;
out:
- if (nolock) {
- if (trans)
- btrfs_end_transaction_nolock(trans, root);
- } else {
+ if (root != root->fs_info->tree_root)
btrfs_delalloc_release_metadata(inode, ordered_extent->len);
- if (trans)
+ if (trans) {
+ if (nolock)
+ btrfs_end_transaction_nolock(trans, root);
+ else
btrfs_end_transaction(trans, root);
}
return btrfs_finish_ordered_io(page->mapping->host, start, end);
}
-/*
- * When IO fails, either with EIO or csum verification fails, we
- * try other mirrors that might have a good copy of the data. This
- * io_failure_record is used to record state as we go through all the
- * mirrors. If another mirror has good data, the page is set up to date
- * and things continue. If a good mirror can't be found, the original
- * bio end_io callback is called to indicate things have failed.
- */
-struct io_failure_record {
- struct page *page;
- u64 start;
- u64 len;
- u64 logical;
- unsigned long bio_flags;
- int last_mirror;
-};
-
-static int btrfs_io_failed_hook(struct bio *failed_bio,
- struct page *page, u64 start, u64 end,
- struct extent_state *state)
-{
- struct io_failure_record *failrec = NULL;
- u64 private;
- struct extent_map *em;
- struct inode *inode = page->mapping->host;
- struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
- struct bio *bio;
- int num_copies;
- int ret;
- int rw;
- u64 logical;
-
- ret = get_state_private(failure_tree, start, &private);
- if (ret) {
- failrec = kmalloc(sizeof(*failrec), GFP_NOFS);
- if (!failrec)
- return -ENOMEM;
- failrec->start = start;
- failrec->len = end - start + 1;
- failrec->last_mirror = 0;
- failrec->bio_flags = 0;
-
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, start, failrec->len);
- if (em->start > start || em->start + em->len < start) {
- free_extent_map(em);
- em = NULL;
- }
- read_unlock(&em_tree->lock);
-
- if (IS_ERR_OR_NULL(em)) {
- kfree(failrec);
- return -EIO;
- }
- logical = start - em->start;
- logical = em->block_start + logical;
- if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
- logical = em->block_start;
- failrec->bio_flags = EXTENT_BIO_COMPRESSED;
- extent_set_compress_type(&failrec->bio_flags,
- em->compress_type);
- }
- failrec->logical = logical;
- free_extent_map(em);
- set_extent_bits(failure_tree, start, end, EXTENT_LOCKED |
- EXTENT_DIRTY, GFP_NOFS);
- set_state_private(failure_tree, start,
- (u64)(unsigned long)failrec);
- } else {
- failrec = (struct io_failure_record *)(unsigned long)private;
- }
- num_copies = btrfs_num_copies(
- &BTRFS_I(inode)->root->fs_info->mapping_tree,
- failrec->logical, failrec->len);
- failrec->last_mirror++;
- if (!state) {
- spin_lock(&BTRFS_I(inode)->io_tree.lock);
- state = find_first_extent_bit_state(&BTRFS_I(inode)->io_tree,
- failrec->start,
- EXTENT_LOCKED);
- if (state && state->start != failrec->start)
- state = NULL;
- spin_unlock(&BTRFS_I(inode)->io_tree.lock);
- }
- if (!state || failrec->last_mirror > num_copies) {
- set_state_private(failure_tree, failrec->start, 0);
- clear_extent_bits(failure_tree, failrec->start,
- failrec->start + failrec->len - 1,
- EXTENT_LOCKED | EXTENT_DIRTY, GFP_NOFS);
- kfree(failrec);
- return -EIO;
- }
- bio = bio_alloc(GFP_NOFS, 1);
- bio->bi_private = state;
- bio->bi_end_io = failed_bio->bi_end_io;
- bio->bi_sector = failrec->logical >> 9;
- bio->bi_bdev = failed_bio->bi_bdev;
- bio->bi_size = 0;
-
- bio_add_page(bio, page, failrec->len, start - page_offset(page));
- if (failed_bio->bi_rw & REQ_WRITE)
- rw = WRITE;
- else
- rw = READ;
-
- ret = BTRFS_I(inode)->io_tree.ops->submit_bio_hook(inode, rw, bio,
- failrec->last_mirror,
- failrec->bio_flags, 0);
- return ret;
-}
-
-/*
- * each time an IO finishes, we do a fast check in the IO failure tree
- * to see if we need to process or clean up an io_failure_record
- */
-static int btrfs_clean_io_failures(struct inode *inode, u64 start)
-{
- u64 private;
- u64 private_failure;
- struct io_failure_record *failure;
- int ret;
-
- private = 0;
- if (count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private,
- (u64)-1, 1, EXTENT_DIRTY, 0)) {
- ret = get_state_private(&BTRFS_I(inode)->io_failure_tree,
- start, &private_failure);
- if (ret == 0) {
- failure = (struct io_failure_record *)(unsigned long)
- private_failure;
- set_state_private(&BTRFS_I(inode)->io_failure_tree,
- failure->start, 0);
- clear_extent_bits(&BTRFS_I(inode)->io_failure_tree,
- failure->start,
- failure->start + failure->len - 1,
- EXTENT_DIRTY | EXTENT_LOCKED,
- GFP_NOFS);
- kfree(failure);
- }
- }
- return 0;
-}
-
/*
* when reads are done, we need to check csums to verify the data is correct
- * if there's a match, we allow the bio to finish. If not, we go through
- * the io_failure_record routines to find good copies
+ * if there's a match, we allow the bio to finish. If not, the code in
+ * extent_io.c will try to find good copies for us.
*/
static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state)
kunmap_atomic(kaddr, KM_USER0);
good:
- /* if the io failure tree for this inode is non-empty,
- * check to see if we've recovered from a failed IO
- */
- btrfs_clean_io_failures(inode, start);
return 0;
zeroit:
up_read(&root->fs_info->cleanup_work_sem);
}
-/*
- * calculate extra metadata reservation when snapshotting a subvolume
- * contains orphan files.
- */
-void btrfs_orphan_pre_snapshot(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending,
- u64 *bytes_to_reserve)
-{
- struct btrfs_root *root;
- struct btrfs_block_rsv *block_rsv;
- u64 num_bytes;
- int index;
-
- root = pending->root;
- if (!root->orphan_block_rsv || list_empty(&root->orphan_list))
- return;
-
- block_rsv = root->orphan_block_rsv;
-
- /* orphan block reservation for the snapshot */
- num_bytes = block_rsv->size;
-
- /*
- * after the snapshot is created, COWing tree blocks may use more
- * space than it frees. So we should make sure there is enough
- * reserved space.
- */
- index = trans->transid & 0x1;
- if (block_rsv->reserved + block_rsv->freed[index] < block_rsv->size) {
- num_bytes += block_rsv->size -
- (block_rsv->reserved + block_rsv->freed[index]);
- }
-
- *bytes_to_reserve += num_bytes;
-}
-
-void btrfs_orphan_post_snapshot(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending)
-{
- struct btrfs_root *root = pending->root;
- struct btrfs_root *snap = pending->snap;
- struct btrfs_block_rsv *block_rsv;
- u64 num_bytes;
- int index;
- int ret;
-
- if (!root->orphan_block_rsv || list_empty(&root->orphan_list))
- return;
-
- /* refill source subvolume's orphan block reservation */
- block_rsv = root->orphan_block_rsv;
- index = trans->transid & 0x1;
- if (block_rsv->reserved + block_rsv->freed[index] < block_rsv->size) {
- num_bytes = block_rsv->size -
- (block_rsv->reserved + block_rsv->freed[index]);
- ret = btrfs_block_rsv_migrate(&pending->block_rsv,
- root->orphan_block_rsv,
- num_bytes);
- BUG_ON(ret);
- }
-
- /* setup orphan block reservation for the snapshot */
- block_rsv = btrfs_alloc_block_rsv(snap);
- BUG_ON(!block_rsv);
-
- btrfs_add_durable_block_rsv(root->fs_info, block_rsv);
- snap->orphan_block_rsv = block_rsv;
-
- num_bytes = root->orphan_block_rsv->size;
- ret = btrfs_block_rsv_migrate(&pending->block_rsv,
- block_rsv, num_bytes);
- BUG_ON(ret);
-
-#if 0
- /* insert orphan item for the snapshot */
- WARN_ON(!root->orphan_item_inserted);
- ret = btrfs_insert_orphan_item(trans, root->fs_info->tree_root,
- snap->root_key.objectid);
- BUG_ON(ret);
- snap->orphan_item_inserted = 1;
-#endif
-}
-
enum btrfs_orphan_cleanup_state {
ORPHAN_CLEANUP_STARTED = 1,
ORPHAN_CLEANUP_DONE = 2,
}
spin_unlock(&root->orphan_lock);
- if (block_rsv)
- btrfs_add_durable_block_rsv(root->fs_info, block_rsv);
-
/* grab metadata reservation from transaction handle */
if (reserve) {
ret = btrfs_orphan_reserve_metadata(trans, inode);
struct btrfs_key key, found_key;
struct btrfs_trans_handle *trans;
struct inode *inode;
+ u64 last_objectid = 0;
int ret = 0, nr_unlink = 0, nr_truncate = 0;
if (cmpxchg(&root->orphan_cleanup_state, 0, ORPHAN_CLEANUP_STARTED))
* crossing root thing. we store the inode number in the
* offset of the orphan item.
*/
+
+ if (found_key.offset == last_objectid) {
+ printk(KERN_ERR "btrfs: Error removing orphan entry, "
+ "stopping orphan cleanup\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ last_objectid = found_key.offset;
+
found_key.objectid = found_key.offset;
found_key.type = BTRFS_INODE_ITEM_KEY;
found_key.offset = 0;
inode = btrfs_iget(root->fs_info->sb, &found_key, root, NULL);
- if (IS_ERR(inode)) {
- ret = PTR_ERR(inode);
+ ret = PTR_RET(inode);
+ if (ret && ret != -ESTALE)
goto out;
- }
/*
- * add this inode to the orphan list so btrfs_orphan_del does
- * the proper thing when we hit it
+ * Inode is already gone but the orphan item is still there,
+ * kill the orphan item.
*/
- spin_lock(&root->orphan_lock);
- list_add(&BTRFS_I(inode)->i_orphan, &root->orphan_list);
- spin_unlock(&root->orphan_lock);
-
- /*
- * if this is a bad inode, means we actually succeeded in
- * removing the inode, but not the orphan record, which means
- * we need to manually delete the orphan since iput will just
- * do a destroy_inode
- */
- if (is_bad_inode(inode)) {
- trans = btrfs_start_transaction(root, 0);
+ if (ret == -ESTALE) {
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto out;
}
- btrfs_orphan_del(trans, inode);
+ ret = btrfs_del_orphan_item(trans, root,
+ found_key.objectid);
+ BUG_ON(ret);
btrfs_end_transaction(trans, root);
- iput(inode);
continue;
}
+ /*
+ * add this inode to the orphan list so btrfs_orphan_del does
+ * the proper thing when we hit it
+ */
+ spin_lock(&root->orphan_lock);
+ list_add(&BTRFS_I(inode)->i_orphan, &root->orphan_list);
+ spin_unlock(&root->orphan_lock);
+
/* if we have links, this was a truncate, lets do that */
if (inode->i_nlink) {
if (!S_ISREG(inode->i_mode)) {
u64 ino = btrfs_ino(inode);
u64 dir_ino = btrfs_ino(dir);
- trans = btrfs_start_transaction(root, 10);
+ /*
+ * 1 for the possible orphan item
+ * 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);
if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
return trans;
return ERR_PTR(-ENOMEM);
}
- trans = btrfs_start_transaction(root, 0);
+ /* 1 for the orphan item */
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
btrfs_free_path(path);
root->fs_info->enospc_unlink = 0;
err = 0;
out:
btrfs_free_path(path);
+ /* Migrate the orphan reservation over */
+ if (!err)
+ err = btrfs_block_rsv_migrate(trans->block_rsv,
+ &root->fs_info->global_block_rsv,
+ trans->bytes_reserved);
+
if (err) {
btrfs_end_transaction(trans, root);
root->fs_info->enospc_unlink = 0;
struct btrfs_root *root)
{
if (trans->block_rsv == &root->fs_info->global_block_rsv) {
+ btrfs_block_rsv_release(root, trans->block_rsv,
+ trans->bytes_reserved);
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
BUG_ON(!root->fs_info->enospc_unlink);
root->fs_info->enospc_unlink = 0;
}
pgoff_t index = from >> PAGE_CACHE_SHIFT;
unsigned offset = from & (PAGE_CACHE_SIZE-1);
struct page *page;
+ gfp_t mask = btrfs_alloc_write_mask(mapping);
int ret = 0;
u64 page_start;
u64 page_end;
ret = -ENOMEM;
again:
- page = find_or_create_page(mapping, index, GFP_NOFS);
+ page = find_or_create_page(mapping, index, mask);
if (!page) {
btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
goto out;
{
struct btrfs_trans_handle *trans;
struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_block_rsv *rsv, *global_rsv;
+ u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
unsigned long nr;
int ret;
goto no_delete;
}
+ rsv = btrfs_alloc_block_rsv(root);
+ if (!rsv) {
+ btrfs_orphan_del(NULL, inode);
+ goto no_delete;
+ }
+ rsv->size = min_size;
+ global_rsv = &root->fs_info->global_block_rsv;
+
btrfs_i_size_write(inode, 0);
+ /*
+ * This is a bit simpler than btrfs_truncate since
+ *
+ * 1) We've already reserved our space for our orphan item in the
+ * unlink.
+ * 2) We're going to delete the inode item, so we don't need to update
+ * it at all.
+ *
+ * So we just need to reserve some slack space in case we add bytes when
+ * doing the truncate.
+ */
while (1) {
- trans = btrfs_join_transaction(root);
- BUG_ON(IS_ERR(trans));
- trans->block_rsv = root->orphan_block_rsv;
+ ret = btrfs_block_rsv_refill(root, rsv, min_size);
+
+ /*
+ * Try and steal from the global reserve since we will
+ * likely not use this space anyway, we want to try as
+ * hard as possible to get this to work.
+ */
+ if (ret)
+ ret = btrfs_block_rsv_migrate(global_rsv, rsv, min_size);
- ret = btrfs_block_rsv_check(trans, root,
- root->orphan_block_rsv, 0, 5);
if (ret) {
- BUG_ON(ret != -EAGAIN);
- ret = btrfs_commit_transaction(trans, root);
- BUG_ON(ret);
- continue;
+ printk(KERN_WARNING "Could not get space for a "
+ "delete, will truncate on mount %d\n", ret);
+ btrfs_orphan_del(NULL, inode);
+ btrfs_free_block_rsv(root, rsv);
+ goto no_delete;
}
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ btrfs_orphan_del(NULL, inode);
+ btrfs_free_block_rsv(root, rsv);
+ goto no_delete;
+ }
+
+ trans->block_rsv = rsv;
+
ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
if (ret != -EAGAIN)
break;
btrfs_end_transaction(trans, root);
trans = NULL;
btrfs_btree_balance_dirty(root, nr);
-
}
+ btrfs_free_block_rsv(root, rsv);
+
if (ret == 0) {
+ trans->block_rsv = root->orphan_block_rsv;
ret = btrfs_orphan_del(trans, inode);
BUG_ON(ret);
}
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
if (!(root == root->fs_info->tree_root ||
root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID))
btrfs_return_ino(root, btrfs_ino(inode));
if (test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) {
ret = btrfs_ordered_update_i_size(inode, 0, ordered);
if (!ret)
- ret = btrfs_update_inode(trans, root, inode);
- err = ret;
+ err = btrfs_update_inode(trans, root, inode);
goto out;
}
{
struct extent_io_tree *tree;
tree = &BTRFS_I(page->mapping->host)->io_tree;
- return extent_read_full_page(tree, page, btrfs_get_extent);
+ return extent_read_full_page(tree, page, btrfs_get_extent, 0);
}
static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
struct btrfs_trans_handle *trans;
unsigned long nr;
u64 mask = root->sectorsize - 1;
+ u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
ret = btrfs_truncate_page(inode->i_mapping, inode->i_size);
if (ret)
rsv = btrfs_alloc_block_rsv(root);
if (!rsv)
return -ENOMEM;
- btrfs_add_durable_block_rsv(root->fs_info, rsv);
+ rsv->size = min_size;
+ /*
+ * 1 for the truncate slack space
+ * 1 for the orphan item we're going to add
+ * 1 for the orphan item deletion
+ * 1 for updating the inode.
+ */
trans = btrfs_start_transaction(root, 4);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
}
- /*
- * Reserve space for the truncate process. Truncate should be adding
- * space, but if there are snapshots it may end up using space.
- */
- ret = btrfs_truncate_reserve_metadata(trans, root, rsv);
+ /* Migrate the slack space for the truncate to our reserve */
+ ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
+ min_size);
BUG_ON(ret);
ret = btrfs_orphan_add(trans, inode);
goto out;
}
- nr = trans->blocks_used;
- btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
-
- /*
- * Ok so we've already migrated our bytes over for the truncate, so here
- * just reserve the one slot we need for updating the inode.
- */
- trans = btrfs_start_transaction(root, 1);
- if (IS_ERR(trans)) {
- err = PTR_ERR(trans);
- goto out;
- }
- trans->block_rsv = rsv;
-
/*
* setattr is responsible for setting the ordered_data_close flag,
* but that is only tested during the last file release. That
btrfs_add_ordered_operation(trans, root, inode);
while (1) {
+ ret = btrfs_block_rsv_refill(root, rsv, min_size);
+ if (ret) {
+ /*
+ * This can only happen with the original transaction we
+ * started above, every other time we shouldn't have a
+ * transaction started yet.
+ */
+ if (ret == -EAGAIN)
+ goto end_trans;
+ err = ret;
+ break;
+ }
+
if (!trans) {
- trans = btrfs_start_transaction(root, 3);
+ /* Just need the 1 for updating the inode */
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
}
-
- ret = btrfs_truncate_reserve_metadata(trans, root,
- rsv);
- BUG_ON(ret);
-
- trans->block_rsv = rsv;
}
+ trans->block_rsv = rsv;
+
ret = btrfs_truncate_inode_items(trans, root, inode,
inode->i_size,
BTRFS_EXTENT_DATA_KEY);
err = ret;
break;
}
-
+end_trans:
nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
trans = NULL;
ei->last_sub_trans = 0;
ei->logged_trans = 0;
ei->delalloc_bytes = 0;
- ei->reserved_bytes = 0;
ei->disk_i_size = 0;
ei->flags = 0;
+ ei->csum_bytes = 0;
ei->index_cnt = (u64)-1;
ei->last_unlink_trans = 0;
WARN_ON(inode->i_data.nrpages);
WARN_ON(BTRFS_I(inode)->outstanding_extents);
WARN_ON(BTRFS_I(inode)->reserved_extents);
+ WARN_ON(BTRFS_I(inode)->delalloc_bytes);
+ WARN_ON(BTRFS_I(inode)->csum_bytes);
/*
* This can happen where we create an inode, but somebody else also
.readpage_end_io_hook = btrfs_readpage_end_io_hook,
.writepage_end_io_hook = btrfs_writepage_end_io_hook,
.writepage_start_hook = btrfs_writepage_start_hook,
- .readpage_io_failed_hook = btrfs_io_failed_hook,
.set_bit_hook = btrfs_set_bit_hook,
.clear_bit_hook = btrfs_clear_bit_hook,
.merge_extent_hook = btrfs_merge_extent_hook,
git.openpandora.org / pandora-kernel.git / blobdiff