diff options
Diffstat (limited to 'fs/btrfs/disk-io.c')
| -rw-r--r-- | fs/btrfs/disk-io.c | 2343 |
1 files changed, 2343 insertions, 0 deletions
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c new file mode 100644 index 00000000000..81a313874ae --- /dev/null +++ b/fs/btrfs/disk-io.c @@ -0,0 +1,2343 @@ +/* + * 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. + */ + +#include <linux/version.h> +#include <linux/fs.h> +#include <linux/blkdev.h> +#include <linux/scatterlist.h> +#include <linux/swap.h> +#include <linux/radix-tree.h> +#include <linux/writeback.h> +#include <linux/buffer_head.h> +#include <linux/workqueue.h> +#include <linux/kthread.h> +#include <linux/freezer.h> +#include "compat.h" +#include "crc32c.h" +#include "ctree.h" +#include "disk-io.h" +#include "transaction.h" +#include "btrfs_inode.h" +#include "volumes.h" +#include "print-tree.h" +#include "async-thread.h" +#include "locking.h" +#include "ref-cache.h" +#include "tree-log.h" + +static struct extent_io_ops btree_extent_io_ops; +static void end_workqueue_fn(struct btrfs_work *work); + +/* + * end_io_wq structs are used to do processing in task context when an IO is + * complete. This is used during reads to verify checksums, and it is used + * by writes to insert metadata for new file extents after IO is complete. + */ +struct end_io_wq { + struct bio *bio; + bio_end_io_t *end_io; + void *private; + struct btrfs_fs_info *info; + int error; + int metadata; + struct list_head list; + struct btrfs_work work; +}; + +/* + * async submit bios are used to offload expensive checksumming + * onto the worker threads. They checksum file and metadata bios + * just before they are sent down the IO stack. + */ +struct async_submit_bio { + struct inode *inode; + struct bio *bio; + struct list_head list; + extent_submit_bio_hook_t *submit_bio_start; + extent_submit_bio_hook_t *submit_bio_done; + int rw; + int mirror_num; + unsigned long bio_flags; + struct btrfs_work work; +}; + +/* + * extents on the btree inode are pretty simple, there's one extent + * that covers the entire device + */ +static struct extent_map *btree_get_extent(struct inode *inode, + struct page *page, size_t page_offset, u64 start, u64 len, + int create) +{ + struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; + struct extent_map *em; + int ret; + + spin_lock(&em_tree->lock); + em = lookup_extent_mapping(em_tree, start, len); + if (em) { + em->bdev = + BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev; + spin_unlock(&em_tree->lock); + goto out; + } + spin_unlock(&em_tree->lock); + + em = alloc_extent_map(GFP_NOFS); + if (!em) { + em = ERR_PTR(-ENOMEM); + goto out; + } + em->start = 0; + em->len = (u64)-1; + em->block_len = (u64)-1; + em->block_start = 0; + em->bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev; + + spin_lock(&em_tree->lock); + ret = add_extent_mapping(em_tree, em); + if (ret == -EEXIST) { + u64 failed_start = em->start; + u64 failed_len = em->len; + + free_extent_map(em); + em = lookup_extent_mapping(em_tree, start, len); + if (em) { + ret = 0; + } else { + em = lookup_extent_mapping(em_tree, failed_start, + failed_len); + ret = -EIO; + } + } else if (ret) { + free_extent_map(em); + em = NULL; + } + spin_unlock(&em_tree->lock); + + if (ret) + em = ERR_PTR(ret); +out: + return em; +} + +u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len) +{ + return btrfs_crc32c(seed, data, len); +} + +void btrfs_csum_final(u32 crc, char *result) +{ + *(__le32 *)result = ~cpu_to_le32(crc); +} + +/* + * compute the csum for a btree block, and either verify it or write it + * into the csum field of the block. + */ +static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf, + int verify) +{ + u16 csum_size = + btrfs_super_csum_size(&root->fs_info->super_copy); + char *result = NULL; + unsigned long len; + unsigned long cur_len; + unsigned long offset = BTRFS_CSUM_SIZE; + char *map_token = NULL; + char *kaddr; + unsigned long map_start; + unsigned long map_len; + int err; + u32 crc = ~(u32)0; + unsigned long inline_result; + + len = buf->len - offset; + while (len > 0) { + err = map_private_extent_buffer(buf, offset, 32, + &map_token, &kaddr, + &map_start, &map_len, KM_USER0); + if (err) + return 1; + cur_len = min(len, map_len - (offset - map_start)); + crc = btrfs_csum_data(root, kaddr + offset - map_start, + crc, cur_len); + len -= cur_len; + offset += cur_len; + unmap_extent_buffer(buf, map_token, KM_USER0); + } + if (csum_size > sizeof(inline_result)) { + result = kzalloc(csum_size * sizeof(char), GFP_NOFS); + if (!result) + return 1; + } else { + result = (char *)&inline_result; + } + + btrfs_csum_final(crc, result); + + if (verify) { + if (memcmp_extent_buffer(buf, result, 0, csum_size)) { + u32 val; + u32 found = 0; + memcpy(&found, result, csum_size); + + read_extent_buffer(buf, &val, 0, csum_size); + printk(KERN_INFO "btrfs: %s checksum verify failed " + "on %llu wanted %X found %X level %d\n", + root->fs_info->sb->s_id, + buf->start, val, found, btrfs_header_level(buf)); + if (result != (char *)&inline_result) + kfree(result); + return 1; + } + } else { + write_extent_buffer(buf, result, 0, csum_size); + } + if (result != (char *)&inline_result) + kfree(result); + return 0; +} + +/* + * we can't consider a given block up to date unless the transid of the + * block matches the transid in the parent node's pointer. This is how we + * detect blocks that either didn't get written at all or got written + * in the wrong place. + */ +static int verify_parent_transid(struct extent_io_tree *io_tree, + struct extent_buffer *eb, u64 parent_transid) +{ + int ret; + + if (!parent_transid || btrfs_header_generation(eb) == parent_transid) + return 0; + + lock_extent(io_tree, eb->start, eb->start + eb->len - 1, GFP_NOFS); + if (extent_buffer_uptodate(io_tree, eb) && + btrfs_header_generation(eb) == parent_transid) { + ret = 0; + goto out; + } + printk("parent transid verify failed on %llu wanted %llu found %llu\n", + (unsigned long long)eb->start, + (unsigned long long)parent_transid, + (unsigned long long)btrfs_header_generation(eb)); + ret = 1; + clear_extent_buffer_uptodate(io_tree, eb); +out: + unlock_extent(io_tree, eb->start, eb->start + eb->len - 1, + GFP_NOFS); + 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. + */ +static int btree_read_extent_buffer_pages(struct btrfs_root *root, + struct extent_buffer *eb, + u64 start, u64 parent_transid) +{ + struct extent_io_tree *io_tree; + int ret; + int num_copies = 0; + int mirror_num = 0; + + io_tree = &BTRFS_I(root->fs_info->btree_inode)->io_tree; + while (1) { + ret = read_extent_buffer_pages(io_tree, eb, start, 1, + btree_get_extent, mirror_num); + if (!ret && + !verify_parent_transid(io_tree, eb, parent_transid)) + return ret; + + num_copies = btrfs_num_copies(&root->fs_info->mapping_tree, + eb->start, eb->len); + if (num_copies == 1) + return ret; + + mirror_num++; + if (mirror_num > num_copies) + return ret; + } + return -EIO; +} + +/* + * checksum a dirty tree block before IO. This has extra checks to make sure + * we only fill in the checksum field in the first page of a multi-page block + */ + +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 found_start; + int found_level; + unsigned long len; + struct extent_buffer *eb; + int ret; + + tree = &BTRFS_I(page->mapping->host)->io_tree; + + if (page->private == EXTENT_PAGE_PRIVATE) + goto out; + if (!page->private) + goto out; + len = page->private >> 2; + WARN_ON(len == 0); + + eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS); + ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE, + btrfs_header_generation(eb)); + BUG_ON(ret); + found_start = btrfs_header_bytenr(eb); + if (found_start != start) { + WARN_ON(1); + goto err; + } + if (eb->first_page != page) { + WARN_ON(1); + goto err; + } + if (!PageUptodate(page)) { + WARN_ON(1); + goto err; + } + found_level = btrfs_header_level(eb); + + csum_tree_block(root, eb, 0); +err: + free_extent_buffer(eb); +out: + return 0; +} + +static int check_tree_block_fsid(struct btrfs_root *root, + struct extent_buffer *eb) +{ + struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices; + u8 fsid[BTRFS_UUID_SIZE]; + int ret = 1; + + read_extent_buffer(eb, fsid, (unsigned long)btrfs_header_fsid(eb), + BTRFS_FSID_SIZE); + while (fs_devices) { + if (!memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE)) { + ret = 0; + break; + } + fs_devices = fs_devices->seed; + } + return ret; +} + +static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end, + struct extent_state *state) +{ + struct extent_io_tree *tree; + u64 found_start; + int found_level; + unsigned long len; + struct extent_buffer *eb; + struct btrfs_root *root = BTRFS_I(page->mapping->host)->root; + int ret = 0; + + tree = &BTRFS_I(page->mapping->host)->io_tree; + if (page->private == EXTENT_PAGE_PRIVATE) + goto out; + if (!page->private) + goto out; + + len = page->private >> 2; + WARN_ON(len == 0); + + eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS); + + found_start = btrfs_header_bytenr(eb); + if (found_start != start) { + printk(KERN_INFO "btrfs bad tree block start %llu %llu\n", + (unsigned long long)found_start, + (unsigned long long)eb->start); + ret = -EIO; + goto err; + } + if (eb->first_page != page) { + printk(KERN_INFO "btrfs bad first page %lu %lu\n", + eb->first_page->index, page->index); + WARN_ON(1); + ret = -EIO; + goto err; + } + if (check_tree_block_fsid(root, eb)) { + printk(KERN_INFO "btrfs bad fsid on block %llu\n", + (unsigned long long)eb->start); + ret = -EIO; + goto err; + } + found_level = btrfs_header_level(eb); + + ret = csum_tree_block(root, eb, 1); + if (ret) + ret = -EIO; + + end = min_t(u64, eb->len, PAGE_CACHE_SIZE); + end = eb->start + end - 1; +err: + free_extent_buffer(eb); +out: + return ret; +} + +static void end_workqueue_bio(struct bio *bio, int err) +{ + struct end_io_wq *end_io_wq = bio->bi_private; + struct btrfs_fs_info *fs_info; + + fs_info = end_io_wq->info; + end_io_wq->error = err; + end_io_wq->work.func = end_workqueue_fn; + end_io_wq->work.flags = 0; + + if (bio->bi_rw & (1 << BIO_RW)) { + if (end_io_wq->metadata) + btrfs_queue_worker(&fs_info->endio_meta_write_workers, + &end_io_wq->work); + else + btrfs_queue_worker(&fs_info->endio_write_workers, + &end_io_wq->work); + } else { + if (end_io_wq->metadata) + btrfs_queue_worker(&fs_info->endio_meta_workers, + &end_io_wq->work); + else + btrfs_queue_worker(&fs_info->endio_workers, + &end_io_wq->work); + } +} + +int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio, + int metadata) +{ + struct end_io_wq *end_io_wq; + end_io_wq = kmalloc(sizeof(*end_io_wq), GFP_NOFS); + if (!end_io_wq) + return -ENOMEM; + + end_io_wq->private = bio->bi_private; + end_io_wq->end_io = bio->bi_end_io; + end_io_wq->info = info; + end_io_wq->error = 0; + end_io_wq->bio = bio; + end_io_wq->metadata = metadata; + + bio->bi_private = end_io_wq; + bio->bi_end_io = end_workqueue_bio; + return 0; +} + +unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info) +{ + unsigned long limit = min_t(unsigned long, + info->workers.max_workers, + info->fs_devices->open_devices); + return 256 * limit; +} + +int btrfs_congested_async(struct btrfs_fs_info *info, int iodone) +{ + return atomic_read(&info->nr_async_bios) > + btrfs_async_submit_limit(info); +} + +static void run_one_async_start(struct btrfs_work *work) +{ + struct btrfs_fs_info *fs_info; + struct async_submit_bio *async; + + async = container_of(work, struct async_submit_bio, work); + fs_info = BTRFS_I(async->inode)->root->fs_info; + async->submit_bio_start(async->inode, async->rw, async->bio, + async->mirror_num, async->bio_flags); +} + +static void run_one_async_done(struct btrfs_work *work) +{ + struct btrfs_fs_info *fs_info; + struct async_submit_bio *async; + int limit; + + async = container_of(work, struct async_submit_bio, work); + fs_info = BTRFS_I(async->inode)->root->fs_info; + + limit = btrfs_async_submit_limit(fs_info); + limit = limit * 2 / 3; + + atomic_dec(&fs_info->nr_async_submits); + + if (atomic_read(&fs_info->nr_async_submits) < limit && + waitqueue_active(&fs_info->async_submit_wait)) + wake_up(&fs_info->async_submit_wait); + + async->submit_bio_done(async->inode, async->rw, async->bio, + async->mirror_num, async->bio_flags); +} + +static void run_one_async_free(struct btrfs_work *work) +{ + struct async_submit_bio *async; + + async = container_of(work, struct async_submit_bio, work); + kfree(async); +} + +int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, + int rw, struct bio *bio, int mirror_num, + unsigned long bio_flags, + extent_submit_bio_hook_t *submit_bio_start, + extent_submit_bio_hook_t *submit_bio_done) +{ + struct async_submit_bio *async; + + async = kmalloc(sizeof(*async), GFP_NOFS); + if (!async) + return -ENOMEM; + + async->inode = inode; + async->rw = rw; + async->bio = bio; + async->mirror_num = mirror_num; + async->submit_bio_start = submit_bio_start; + async->submit_bio_done = submit_bio_done; + + async->work.func = run_one_async_start; + async->work.ordered_func = run_one_async_done; + async->work.ordered_free = run_one_async_free; + + async->work.flags = 0; + async->bio_flags = bio_flags; + + atomic_inc(&fs_info->nr_async_submits); + btrfs_queue_worker(&fs_info->workers, &async->work); +#if 0 + int limit = btrfs_async_submit_limit(fs_info); + if (atomic_read(&fs_info->nr_async_submits) > limit) { + wait_event_timeout(fs_info->async_submit_wait, + (atomic_read(&fs_info->nr_async_submits) < limit), + HZ/10); + + wait_event_timeout(fs_info->async_submit_wait, + (atomic_read(&fs_info->nr_async_bios) < limit), + HZ/10); + } +#endif + while (atomic_read(&fs_info->async_submit_draining) && + atomic_read(&fs_info->nr_async_submits)) { + wait_event(fs_info->async_submit_wait, + (atomic_read(&fs_info->nr_async_submits) == 0)); + } + + return 0; +} + +static int btree_csum_one_bio(struct bio *bio) +{ + struct bio_vec *bvec = bio->bi_io_vec; + int bio_index = 0; + struct btrfs_root *root; + + WARN_ON(bio->bi_vcnt <= 0); + while (bio_index < bio->bi_vcnt) { + root = BTRFS_I(bvec->bv_page->mapping->host)->root; + csum_dirty_buffer(root, bvec->bv_page); + bio_index++; + bvec++; + } + return 0; +} + +static int __btree_submit_bio_start(struct inode *inode, int rw, + struct bio *bio, int mirror_num, + unsigned long bio_flags) +{ + /* + * when we're called for a write, we're already in the async + * submission context. Just jump into btrfs_map_bio + */ + btree_csum_one_bio(bio); + return 0; +} + +static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio, + int mirror_num, unsigned long bio_flags) +{ + /* + * when we're called for a write, we're already in the async + * submission context. Just jump into btrfs_map_bio + */ + return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1); +} + +static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, + int mirror_num, unsigned long bio_flags) +{ + int ret; + + ret = btrfs_bio_wq_end_io(BTRFS_I(inode)->root->fs_info, + bio, 1); + BUG_ON(ret); + + if (!(rw & (1 << BIO_RW))) { + /* + * called for a read, do the setup so that checksum validation + * can happen in the async kernel threads + */ + return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, + mirror_num, 0); + } + /* + * kthread helpers are used to submit writes so that checksumming + * can happen in parallel across all CPUs + */ + return btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info, + inode, rw, bio, mirror_num, 0, + __btree_submit_bio_start, + __btree_submit_bio_done); +} + +static int btree_writepage(struct page *page, struct writeback_control *wbc) +{ + struct extent_io_tree *tree; + tree = &BTRFS_I(page->mapping->host)->io_tree; + + if (current->flags & PF_MEMALLOC) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + return extent_write_full_page(tree, page, btree_get_extent, wbc); +} + +static int btree_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct extent_io_tree *tree; + tree = &BTRFS_I(mapping->host)->io_tree; + if (wbc->sync_mode == WB_SYNC_NONE) { + u64 num_dirty; + u64 start = 0; + unsigned long thresh = 32 * 1024 * 1024; + + if (wbc->for_kupdate) + return 0; + + num_dirty = count_range_bits(tree, &start, (u64)-1, + thresh, EXTENT_DIRTY); + if (num_dirty < thresh) + return 0; + } + return extent_writepages(tree, mapping, btree_get_extent, wbc); +} + +static int btree_readpage(struct file *file, struct page *page) +{ + struct extent_io_tree *tree; + tree = &BTRFS_I(page->mapping->host)->io_tree; + return extent_read_full_page(tree, page, btree_get_extent); +} + +static int btree_releasepage(struct page *page, gfp_t gfp_flags) +{ + struct extent_io_tree *tree; + struct extent_map_tree *map; + int ret; + + if (PageWriteback(page) || PageDirty(page)) + return 0; + + tree = &BTRFS_I(page->mapping->host)->io_tree; + map = &BTRFS_I(page->mapping->host)->extent_tree; + + ret = try_release_extent_state(map, tree, page, gfp_flags); + if (!ret) + return 0; + + ret = try_release_extent_buffer(tree, page); + if (ret == 1) { + ClearPagePrivate(page); + set_page_private(page, 0); + page_cache_release(page); + } + + return ret; +} + +static void btree_invalidatepage(struct page *page, unsigned long offset) +{ + struct extent_io_tree *tree; + tree = &BTRFS_I(page->mapping->host)->io_tree; + extent_invalidatepage(tree, page, offset); + btree_releasepage(page, GFP_NOFS); + if (PagePrivate(page)) { + printk(KERN_WARNING "btrfs warning page private not zero " + "on page %llu\n", (unsigned long long)page_offset(page)); + ClearPagePrivate(page); + set_page_private(page, 0); + page_cache_release(page); + } +} + +#if 0 +static int btree_writepage(struct page *page, struct writeback_control *wbc) +{ + struct buffer_head *bh; + struct btrfs_root *root = BTRFS_I(page->mapping->host)->root; + struct buffer_head *head; + if (!page_has_buffers(page)) { + create_empty_buffers(page, root->fs_info->sb->s_blocksize, + (1 << BH_Dirty)|(1 << BH_Uptodate)); + } + head = page_buffers(page); + bh = head; + do { + if (buffer_dirty(bh)) + csum_tree_block(root, bh, 0); + bh = bh->b_this_page; + } while (bh != head); + return block_write_full_page(page, btree_get_block, wbc); +} +#endif + +static struct address_space_operations btree_aops = { + .readpage = btree_readpage, + .writepage = btree_writepage, + .writepages = btree_writepages, + .releasepage = btree_releasepage, + .invalidatepage = btree_invalidatepage, + .sync_page = block_sync_page, +}; + +int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize, + u64 parent_transid) +{ + struct extent_buffer *buf = NULL; + struct inode *btree_inode = root->fs_info->btree_inode; + int ret = 0; + + buf = btrfs_find_create_tree_block(root, bytenr, blocksize); + if (!buf) + return 0; + read_extent_buffer_pages(&BTRFS_I(btree_inode)->io_tree, + buf, 0, 0, btree_get_extent, 0); + free_extent_buffer(buf); + return ret; +} + +struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root, + u64 bytenr, u32 blocksize) +{ + struct inode *btree_inode = root->fs_info->btree_inode; + struct extent_buffer *eb; + eb = find_extent_buffer(&BTRFS_I(btree_inode)->io_tree, + bytenr, blocksize, GFP_NOFS); + return eb; +} + +struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root, + u64 bytenr, u32 blocksize) +{ + struct inode *btree_inode = root->fs_info->btree_inode; + struct extent_buffer *eb; + + eb = alloc_extent_buffer(&BTRFS_I(btree_inode)->io_tree, + bytenr, blocksize, NULL, GFP_NOFS); + return eb; +} + + +int btrfs_write_tree_block(struct extent_buffer *buf) +{ + return btrfs_fdatawrite_range(buf->first_page->mapping, buf->start, + buf->start + buf->len - 1, WB_SYNC_ALL); +} + +int btrfs_wait_tree_block_writeback(struct extent_buffer *buf) +{ + return btrfs_wait_on_page_writeback_range(buf->first_page->mapping, + buf->start, buf->start + buf->len - 1); +} + +struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr, + u32 blocksize, u64 parent_transid) +{ + struct extent_buffer *buf = NULL; + struct inode *btree_inode = root->fs_info->btree_inode; + struct extent_io_tree *io_tree; + int ret; + + io_tree = &BTRFS_I(btree_inode)->io_tree; + + buf = btrfs_find_create_tree_block(root, bytenr, blocksize); + if (!buf) + return NULL; + + ret = btree_read_extent_buffer_pages(root, buf, 0, parent_transid); + + if (ret == 0) + buf->flags |= EXTENT_UPTODATE; + else + WARN_ON(1); + return buf; + +} + +int clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, + struct extent_buffer *buf) +{ + struct inode *btree_inode = root->fs_info->btree_inode; + if (btrfs_header_generation(buf) == + root->fs_info->running_transaction->transid) { + WARN_ON(!btrfs_tree_locked(buf)); + clear_extent_buffer_dirty(&BTRFS_I(btree_inode)->io_tree, + buf); + } + return 0; +} + +static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize, + u32 stripesize, struct btrfs_root *root, + struct btrfs_fs_info *fs_info, + u64 objectid) +{ + root->node = NULL; + root->commit_root = NULL; + root->ref_tree = NULL; + root->sectorsize = sectorsize; + root->nodesize = nodesize; + root->leafsize = leafsize; + root->stripesize = stripesize; + root->ref_cows = 0; + root->track_dirty = 0; + + root->fs_info = fs_info; + root->objectid = objectid; + root->last_trans = 0; + root->highest_inode = 0; + root->last_inode_alloc = 0; + root->name = NULL; + root->in_sysfs = 0; + + INIT_LIST_HEAD(&root->dirty_list); + INIT_LIST_HEAD(&root->orphan_list); + INIT_LIST_HEAD(&root->dead_list); + spin_lock_init(&root->node_lock); + spin_lock_init(&root->list_lock); + mutex_init(&root->objectid_mutex); + mutex_init(&root->log_mutex); + extent_io_tree_init(&root->dirty_log_pages, + fs_info->btree_inode->i_mapping, GFP_NOFS); + + btrfs_leaf_ref_tree_init(&root->ref_tree_struct); + root->ref_tree = &root->ref_tree_struct; + + memset(&root->root_key, 0, sizeof(root->root_key)); + memset(&root->root_item, 0, sizeof(root->root_item)); + memset(&root->defrag_progress, 0, sizeof(root->defrag_progress)); + memset(&root->root_kobj, 0, sizeof(root->root_kobj)); + root->defrag_trans_start = fs_info->generation; + init_completion(&root->kobj_unregister); + root->defrag_running = 0; + root->defrag_level = 0; + root->root_key.objectid = objectid; + root->anon_super.s_root = NULL; + root->anon_super.s_dev = 0; + INIT_LIST_HEAD(&root->anon_super.s_list); + INIT_LIST_HEAD(&root->anon_super.s_instances); + init_rwsem(&root->anon_super.s_umount); + + return 0; +} + +static int find_and_setup_root(struct btrfs_root *tree_root, + struct btrfs_fs_info *fs_info, + u64 objectid, + struct btrfs_root *root) +{ + int ret; + u32 blocksize; + u64 generation; + + __setup_root(tree_root->nodesize, tree_root->leafsize, + tree_root->sectorsize, tree_root->stripesize, + root, fs_info, objectid); + ret = btrfs_find_last_root(tree_root, objectid, + &root->root_item, &root->root_key); + BUG_ON(ret); + + generation = btrfs_root_generation(&root->root_item); + 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); + BUG_ON(!root->node); + return 0; +} + +int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info) +{ + struct extent_buffer *eb; + struct btrfs_root *log_root_tree = fs_info->log_root_tree; + u64 start = 0; + u64 end = 0; + int ret; + + if (!log_root_tree) + return 0; + + while (1) { + ret = find_first_extent_bit(&log_root_tree->dirty_log_pages, + 0, &start, &end, EXTENT_DIRTY); + if (ret) + break; + + clear_extent_dirty(&log_root_tree->dirty_log_pages, + start, end, GFP_NOFS); + } + eb = fs_info->log_root_tree->node; + + WARN_ON(btrfs_header_level(eb) != 0); + WARN_ON(btrfs_header_nritems(eb) != 0); + + ret = btrfs_free_reserved_extent(fs_info->tree_root, + eb->start, eb->len); + BUG_ON(ret); + + free_extent_buffer(eb); + kfree(fs_info->log_root_tree); + fs_info->log_root_tree = NULL; + return 0; +} + +int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info) +{ + struct btrfs_root *root; + struct btrfs_root *tree_root = fs_info->tree_root; + + root = kzalloc(sizeof(*root), GFP_NOFS); + if (!root) + return -ENOMEM; + + __setup_root(tree_root->nodesize, tree_root->leafsize, + tree_root->sectorsize, tree_root->stripesize, + root, fs_info, BTRFS_TREE_LOG_OBJECTID); + + root->root_key.objectid = BTRFS_TREE_LOG_OBJECTID; + root->root_key.type = BTRFS_ROOT_ITEM_KEY; + root->root_key.offset = BTRFS_TREE_LOG_OBJECTID; + root->ref_cows = 0; + + root->node = btrfs_alloc_free_block(trans, root, root->leafsize, + 0, BTRFS_TREE_LOG_OBJECTID, + trans->transid, 0, 0, 0); + + btrfs_set_header_nritems(root->node, 0); + btrfs_set_header_level(root->node, 0); + btrfs_set_header_bytenr(root->node, root->node->start); + btrfs_set_header_generation(root->node, trans->transid); + btrfs_set_header_owner(root->node, BTRFS_TREE_LOG_OBJECTID); + + write_extent_buffer(root->node, root->fs_info->fsid, + (unsigned long)btrfs_header_fsid(root->node), + BTRFS_FSID_SIZE); + btrfs_mark_buffer_dirty(root->node); + btrfs_tree_unlock(root->node); + fs_info->log_root_tree = root; + return 0; +} + +struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root, + struct btrfs_key *location) +{ + struct btrfs_root *root; + struct btrfs_fs_info *fs_info = tree_root->fs_info; + struct btrfs_path *path; + struct extent_buffer *l; + u64 highest_inode; + u64 generation; + u32 blocksize; + int ret = 0; + + root = kzalloc(sizeof(*root), GFP_NOFS); + if (!root) + return ERR_PTR(-ENOMEM); + if (location->offset == (u64)-1) { + ret = find_and_setup_root(tree_root, fs_info, + location->objectid, root); + if (ret) { + kfree(root); + return ERR_PTR(ret); + } + goto insert; + } + + __setup_root(tree_root->nodesize, tree_root->leafsize, + tree_root->sectorsize, tree_root->stripesize, + root, fs_info, location->objectid); + + path = btrfs_alloc_path(); + BUG_ON(!path); + ret = btrfs_search_slot(NULL, tree_root, location, path, 0, 0); + if (ret != 0) { + if (ret > 0) + ret = -ENOENT; + goto out; + } + l = path->nodes[0]; + read_extent_buffer(l, &root->root_item, + btrfs_item_ptr_offset(l, path->slots[0]), + sizeof(root->root_item)); + memcpy(&root->root_key, location, sizeof(*location)); + ret = 0; +out: + btrfs_release_path(root, path); + btrfs_free_path(path); + if (ret) { + kfree(root); + return ERR_PTR(ret); + } + generation = btrfs_root_generation(&root->root_item); + 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); + BUG_ON(!root->node); +insert: + if (location->objectid != BTRFS_TREE_LOG_OBJECTID) { + root->ref_cows = 1; + ret = btrfs_find_highest_inode(root, &highest_inode); + if (ret == 0) { + root->highest_inode = highest_inode; + root->last_inode_alloc = highest_inode; + } + } + return root; +} + +struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info, + u64 root_objectid) +{ + struct btrfs_root *root; + + if (root_objectid == BTRFS_ROOT_TREE_OBJECTID) + return fs_info->tree_root; + if (root_objectid == BTRFS_EXTENT_TREE_OBJECTID) + return fs_info->extent_root; + + root = radix_tree_lookup(&fs_info->fs_roots_radix, + (unsigned long)root_objectid); + return root; +} + +struct btrfs_root *btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info, + struct btrfs_key *location) +{ + struct btrfs_root *root; + int ret; + + if (location->objectid == BTRFS_ROOT_TREE_OBJECTID) + return fs_info->tree_root; + if (location->objectid == BTRFS_EXTENT_TREE_OBJECTID) + return fs_info->extent_root; + if (location->objectid == BTRFS_CHUNK_TREE_OBJECTID) + return fs_info->chunk_root; + if (location->objectid == BTRFS_DEV_TREE_OBJECTID) + return fs_info->dev_root; + if (location->objectid == BTRFS_CSUM_TREE_OBJECTID) + return fs_info->csum_root; + + root = radix_tree_lookup(&fs_info->fs_roots_radix, + (unsigned long)location->objectid); + if (root) + return root; + + root = btrfs_read_fs_root_no_radix(fs_info->tree_root, location); + if (IS_ERR(root)) + return root; + + set_anon_super(&root->anon_super, NULL); + + ret = radix_tree_insert(&fs_info->fs_roots_radix, + (unsigned long)root->root_key.objectid, + root); + if (ret) { + free_extent_buffer(root->node); + kfree(root); + return ERR_PTR(ret); + } + if (!(fs_info->sb->s_flags & MS_RDONLY)) { + ret = btrfs_find_dead_roots(fs_info->tree_root, + root->root_key.objectid, root); + BUG_ON(ret); + btrfs_orphan_cleanup(root); + } + return root; +} + +struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info, + struct btrfs_key *location, + const char *name, int namelen) +{ + struct btrfs_root *root; + int ret; + + root = btrfs_read_fs_root_no_name(fs_info, location); + if (!root) + return NULL; + + if (root->in_sysfs) + return root; + + ret = btrfs_set_root_name(root, name, namelen); + if (ret) { + free_extent_buffer(root->node); + kfree(root); + return ERR_PTR(ret); + } +#if 0 + ret = btrfs_sysfs_add_root(root); + if (ret) { + free_extent_buffer(root->node); + kfree(root->name); + kfree(root); + return ERR_PTR(ret); + } +#endif + root->in_sysfs = 1; + return root; +} + +static int btrfs_congested_fn(void *congested_data, int bdi_bits) +{ + struct btrfs_fs_info *info = (struct btrfs_fs_info *)congested_data; + int ret = 0; + struct list_head *cur; + struct btrfs_device *device; + struct backing_dev_info *bdi; +#if 0 + if ((bdi_bits & (1 << BDI_write_congested)) && + btrfs_congested_async(info, 0)) + return 1; +#endif + list_for_each(cur, &info->fs_devices->devices) { + device = list_entry(cur, struct btrfs_device, dev_list); + if (!device->bdev) + continue; + bdi = blk_get_backing_dev_info(device->bdev); + if (bdi && bdi_congested(bdi, bdi_bits)) { + ret = 1; + break; + } + } + return ret; +} + +/* + * this unplugs every device on the box, and it is only used when page + * is null + */ +static void __unplug_io_fn(struct backing_dev_info *bdi, struct page *page) +{ + struct list_head *cur; + struct btrfs_device *device; + struct btrfs_fs_info *info; + + info = (struct btrfs_fs_info *)bdi->unplug_io_data; + list_for_each(cur, &info->fs_devices->devices) { + device = list_entry(cur, struct btrfs_device, dev_list); + if (!device->bdev) + continue; + + bdi = blk_get_backing_dev_info(device->bdev); + if (bdi->unplug_io_fn) + bdi->unplug_io_fn(bdi, page); + } +} + +static void btrfs_unplug_io_fn(struct backing_dev_info *bdi, struct page *page) +{ + struct inode *inode; + struct extent |