diff options
Diffstat (limited to 'fs/btrfs/inode.c')
| -rw-r--r-- | fs/btrfs/inode.c | 1888 |
1 files changed, 925 insertions, 963 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8e3a5a266917..5e71f1ea3391 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -30,7 +30,6 @@ #include <linux/mpage.h> #include <linux/swap.h> #include <linux/writeback.h> -#include <linux/statfs.h> #include <linux/compat.h> #include <linux/bit_spinlock.h> #include <linux/xattr.h> @@ -72,6 +71,7 @@ struct btrfs_dio_data { u64 reserve; u64 unsubmitted_oe_range_start; u64 unsubmitted_oe_range_end; + int overwrite; }; static const struct inode_operations btrfs_dir_inode_operations; @@ -109,11 +109,11 @@ static noinline int cow_file_range(struct inode *inode, u64 start, u64 end, u64 delalloc_end, int *page_started, unsigned long *nr_written, int unlock, struct btrfs_dedupe_hash *hash); -static struct extent_map *create_pinned_em(struct inode *inode, u64 start, - u64 len, u64 orig_start, - u64 block_start, u64 block_len, - u64 orig_block_len, u64 ram_bytes, - int type); +static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, + u64 orig_start, u64 block_start, + u64 block_len, u64 orig_block_len, + u64 ram_bytes, int compress_type, + int type); static int btrfs_dirty_inode(struct inode *inode); @@ -167,7 +167,7 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, struct btrfs_key key; size_t datasize; - key.objectid = btrfs_ino(inode); + key.objectid = btrfs_ino(BTRFS_I(inode)); key.offset = start; key.type = BTRFS_EXTENT_DATA_KEY; @@ -250,11 +250,12 @@ static noinline int cow_file_range_inline(struct btrfs_root *root, int compress_type, struct page **compressed_pages) { + struct btrfs_fs_info *fs_info = root->fs_info; 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 = ALIGN(end, root->sectorsize); + u64 aligned_end = ALIGN(end, fs_info->sectorsize); u64 data_len = inline_len; int ret; struct btrfs_path *path; @@ -265,12 +266,12 @@ static noinline int cow_file_range_inline(struct btrfs_root *root, data_len = compressed_size; if (start > 0 || - actual_end > root->sectorsize || - data_len > BTRFS_MAX_INLINE_DATA_SIZE(root) || + actual_end > fs_info->sectorsize || + data_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info) || (!compressed_size && - (actual_end & (root->sectorsize - 1)) == 0) || + (actual_end & (fs_info->sectorsize - 1)) == 0) || end + 1 < isize || - data_len > root->fs_info->max_inline) { + data_len > fs_info->max_inline) { return 1; } @@ -283,7 +284,7 @@ static noinline int cow_file_range_inline(struct btrfs_root *root, btrfs_free_path(path); return PTR_ERR(trans); } - trans->block_rsv = &root->fs_info->delalloc_block_rsv; + trans->block_rsv = &fs_info->delalloc_block_rsv; if (compressed_size && compressed_pages) extent_item_size = btrfs_file_extent_calc_inline_size( @@ -315,8 +316,8 @@ static noinline int cow_file_range_inline(struct btrfs_root *root, } set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); - btrfs_delalloc_release_metadata(inode, end + 1 - start); - btrfs_drop_extent_cache(inode, start, aligned_end - 1, 0); + btrfs_delalloc_release_metadata(BTRFS_I(inode), end + 1 - start); + btrfs_drop_extent_cache(BTRFS_I(inode), start, aligned_end - 1, 0); out: /* * Don't forget to free the reserved space, as for inlined extent @@ -326,7 +327,7 @@ out: */ btrfs_qgroup_free_data(inode, 0, PAGE_SIZE); btrfs_free_path(path); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); return ret; } @@ -373,21 +374,30 @@ static noinline int add_async_extent(struct async_cow *cow, static inline int inode_need_compress(struct inode *inode) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); /* force compress */ - if (btrfs_test_opt(root->fs_info, FORCE_COMPRESS)) + if (btrfs_test_opt(fs_info, FORCE_COMPRESS)) return 1; /* bad compression ratios */ if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS) return 0; - if (btrfs_test_opt(root->fs_info, COMPRESS) || + if (btrfs_test_opt(fs_info, COMPRESS) || BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS || BTRFS_I(inode)->force_compress) return 1; return 0; } +static inline void inode_should_defrag(struct btrfs_inode *inode, + u64 start, u64 end, u64 num_bytes, u64 small_write) +{ + /* If this is a small write inside eof, kick off a defrag */ + if (num_bytes < small_write && + (start > 0 || end + 1 < inode->disk_i_size)) + btrfs_add_inode_defrag(NULL, inode); +} + /* * we create compressed extents in two phases. The first * phase compresses a range of pages that have already been @@ -411,34 +421,32 @@ static noinline void compress_file_range(struct inode *inode, struct async_cow *async_cow, int *num_added) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; u64 num_bytes; - u64 blocksize = root->sectorsize; + u64 blocksize = fs_info->sectorsize; u64 actual_end; u64 isize = i_size_read(inode); int ret = 0; struct page **pages = NULL; unsigned long nr_pages; - unsigned long nr_pages_ret = 0; unsigned long total_compressed = 0; unsigned long total_in = 0; - unsigned long max_compressed = SZ_128K; - unsigned long max_uncompressed = SZ_128K; int i; int will_compress; - int compress_type = root->fs_info->compress_type; + int compress_type = fs_info->compress_type; int redirty = 0; - /* if this is a small write inside eof, kick off a defrag */ - if ((end - start + 1) < SZ_16K && - (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size)) - btrfs_add_inode_defrag(NULL, inode); + inode_should_defrag(BTRFS_I(inode), start, end, end - start + 1, + SZ_16K); actual_end = min_t(u64, isize, end + 1); again: will_compress = 0; nr_pages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1; - nr_pages = min_t(unsigned long, nr_pages, SZ_128K / PAGE_SIZE); + BUILD_BUG_ON((BTRFS_MAX_COMPRESSED % PAGE_SIZE) != 0); + nr_pages = min_t(unsigned long, nr_pages, + BTRFS_MAX_COMPRESSED / PAGE_SIZE); /* * we don't want to send crud past the end of i_size through @@ -463,17 +471,8 @@ again: (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size)) goto cleanup_and_bail_uncompressed; - /* we want to make sure that amount of ram required to uncompress - * an extent is reasonable, so we limit the total size in ram - * of a compressed extent to 128k. This is a crucial number - * because it also controls how easily we can spread reads across - * cpus for decompression. - * - * We also want to make sure the amount of IO required to do - * a random read is reasonably small, so we limit the size of - * a compressed extent to 128k. - */ - total_compressed = min(total_compressed, max_uncompressed); + total_compressed = min_t(unsigned long, total_compressed, + BTRFS_MAX_UNCOMPRESSED); num_bytes = ALIGN(end - start + 1, blocksize); num_bytes = max(blocksize, num_bytes); total_in = 0; @@ -508,16 +507,15 @@ again: redirty = 1; ret = btrfs_compress_pages(compress_type, inode->i_mapping, start, - total_compressed, pages, - nr_pages, &nr_pages_ret, + pages, + &nr_pages, &total_in, - &total_compressed, - max_compressed); + &total_compressed); if (!ret) { unsigned long offset = total_compressed & (PAGE_SIZE - 1); - struct page *page = pages[nr_pages_ret - 1]; + struct page *page = pages[nr_pages - 1]; char *kaddr; /* zero the tail end of the last page, we might be @@ -540,7 +538,7 @@ cont: * to make an uncompressed inline extent. */ ret = cow_file_range_inline(root, inode, start, end, - 0, 0, NULL); + 0, BTRFS_COMPRESS_NONE, NULL); } else { /* try making a compressed inline extent */ ret = cow_file_range_inline(root, inode, start, end, @@ -598,7 +596,7 @@ cont: * will submit them to the elevator. */ add_async_extent(async_cow, start, num_bytes, - total_compressed, pages, nr_pages_ret, + total_compressed, pages, nr_pages, compress_type); if (start + num_bytes < end) { @@ -615,17 +613,17 @@ cont: * the compression code ran but failed to make things smaller, * free any pages it allocated and our page pointer array */ - for (i = 0; i < nr_pages_ret; i++) { + for (i = 0; i < nr_pages; i++) { WARN_ON(pages[i]->mapping); put_page(pages[i]); } kfree(pages); pages = NULL; total_compressed = 0; - nr_pages_ret = 0; + nr_pages = 0; /* flag the file so we don't compress in the future */ - if (!btrfs_test_opt(root->fs_info, FORCE_COMPRESS) && + if (!btrfs_test_opt(fs_info, FORCE_COMPRESS) && !(BTRFS_I(inode)->force_compress)) { BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS; } @@ -651,7 +649,7 @@ cleanup_and_bail_uncompressed: return; free_pages_out: - for (i = 0; i < nr_pages_ret; i++) { + for (i = 0; i < nr_pages; i++) { WARN_ON(pages[i]->mapping); put_page(pages[i]); } @@ -683,12 +681,12 @@ static void free_async_extent_pages(struct async_extent *async_extent) static noinline void submit_compressed_extents(struct inode *inode, struct async_cow *async_cow) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct async_extent *async_extent; u64 alloc_hint = 0; struct btrfs_key ins; struct extent_map *em; struct btrfs_root *root = BTRFS_I(inode)->root; - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; struct extent_io_tree *io_tree; int ret = 0; @@ -776,46 +774,19 @@ retry: * here we're doing allocation and writeback of the * compressed pages */ - btrfs_drop_extent_cache(inode, async_extent->start, - async_extent->start + - async_extent->ram_size - 1, 0); - - em = alloc_extent_map(); - if (!em) { - ret = -ENOMEM; - goto out_free_reserve; - } - em->start = async_extent->start; - em->len = async_extent->ram_size; - em->orig_start = em->start; - em->mod_start = em->start; - em->mod_len = em->len; - - em->block_start = ins.objectid; - em->block_len = ins.offset; - em->orig_block_len = ins.offset; - em->ram_bytes = async_extent->ram_size; - em->bdev = root->fs_info->fs_devices->latest_bdev; - em->compress_type = async_extent->compress_type; - set_bit(EXTENT_FLAG_PINNED, &em->flags); - set_bit(EXTENT_FLAG_COMPRESSED, &em->flags); - em->generation = -1; - - while (1) { - write_lock(&em_tree->lock); - ret = add_extent_mapping(em_tree, em, 1); - write_unlock(&em_tree->lock); - if (ret != -EEXIST) { - free_extent_map(em); - break; - } - btrfs_drop_extent_cache(inode, async_extent->start, - async_extent->start + - async_extent->ram_size - 1, 0); - } - - if (ret) + em = create_io_em(inode, async_extent->start, + async_extent->ram_size, /* len */ + async_extent->start, /* orig_start */ + ins.objectid, /* block_start */ + ins.offset, /* block_len */ + ins.offset, /* orig_block_len */ + async_extent->ram_size, /* ram_bytes */ + async_extent->compress_type, + BTRFS_ORDERED_COMPRESSED); + if (IS_ERR(em)) + /* ret value is not necessary due to void function */ goto out_free_reserve; + free_extent_map(em); ret = btrfs_add_ordered_extent_compress(inode, async_extent->start, @@ -825,12 +796,13 @@ retry: BTRFS_ORDERED_COMPRESSED, async_extent->compress_type); if (ret) { - btrfs_drop_extent_cache(inode, async_extent->start, + btrfs_drop_extent_cache(BTRFS_I(inode), + async_extent->start, async_extent->start + async_extent->ram_size - 1, 0); goto out_free_reserve; } - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); /* * clear dirty, set writeback and unlock the pages. @@ -871,8 +843,8 @@ retry: } return; out_free_reserve: - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); - btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 1); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); out_free: extent_clear_unlock_delalloc(inode, async_extent->start, async_extent->start + @@ -940,19 +912,19 @@ static noinline int cow_file_range(struct inode *inode, int *page_started, unsigned long *nr_written, int unlock, struct btrfs_dedupe_hash *hash) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; u64 alloc_hint = 0; u64 num_bytes; unsigned long ram_size; u64 disk_num_bytes; u64 cur_alloc_size; - u64 blocksize = root->sectorsize; + u64 blocksize = fs_info->sectorsize; struct btrfs_key ins; struct extent_map *em; - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; int ret = 0; - if (btrfs_is_free_space_inode(inode)) { + if (btrfs_is_free_space_inode(BTRFS_I(inode))) { WARN_ON_ONCE(1); ret = -EINVAL; goto out_unlock; @@ -962,15 +934,12 @@ static noinline int cow_file_range(struct inode *inode, num_bytes = max(blocksize, num_bytes); disk_num_bytes = num_bytes; - /* if this is a small write inside eof, kick off defrag */ - if (num_bytes < SZ_64K && - (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size)) - btrfs_add_inode_defrag(NULL, inode); + inode_should_defrag(BTRFS_I(inode), start, end, num_bytes, SZ_64K); if (start == 0) { /* lets try to make an inline extent */ - ret = cow_file_range_inline(root, inode, start, end, 0, 0, - NULL); + ret = cow_file_range_inline(root, inode, start, end, 0, + BTRFS_COMPRESS_NONE, NULL); if (ret == 0) { extent_clear_unlock_delalloc(inode, start, end, delalloc_end, NULL, @@ -990,54 +959,34 @@ static noinline int cow_file_range(struct inode *inode, } BUG_ON(disk_num_bytes > - btrfs_super_total_bytes(root->fs_info->super_copy)); + btrfs_super_total_bytes(fs_info->super_copy)); alloc_hint = get_extent_allocation_hint(inode, start, num_bytes); - btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0); + btrfs_drop_extent_cache(BTRFS_I(inode), start, + start + num_bytes - 1, 0); while (disk_num_bytes > 0) { unsigned long op; cur_alloc_size = disk_num_bytes; ret = btrfs_reserve_extent(root, cur_alloc_size, cur_alloc_size, - root->sectorsize, 0, alloc_hint, + fs_info->sectorsize, 0, alloc_hint, &ins, 1, 1); if (ret < 0) goto out_unlock; - em = alloc_extent_map(); - if (!em) { - ret = -ENOMEM; - goto out_reserve; - } - em->start = start; - em->orig_start = em->start; ram_size = ins.offset; - em->len = ins.offset; - em->mod_start = em->start; - em->mod_len = em->len; - - em->block_start = ins.objectid; - em->block_len = ins.offset; - em->orig_block_len = ins.offset; - em->ram_bytes = ram_size; - em->bdev = 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, 1); - write_unlock(&em_tree->lock); - if (ret != -EEXIST) { - free_extent_map(em); - break; - } - btrfs_drop_extent_cache(inode, start, - start + ram_size - 1, 0); - } - if (ret) + em = create_io_em(inode, start, ins.offset, /* len */ + start, /* orig_start */ + ins.objectid, /* block_start */ + ins.offset, /* block_len */ + ins.offset, /* orig_block_len */ + ram_size, /* ram_bytes */ + BTRFS_COMPRESS_NONE, /* compress_type */ + BTRFS_ORDERED_REGULAR /* type */); + if (IS_ERR(em)) goto out_reserve; + free_extent_map(em); cur_alloc_size = ins.offset; ret = btrfs_add_ordered_extent(inode, start, ins.objectid, @@ -1053,7 +1002,7 @@ static noinline int cow_file_range(struct inode *inode, goto out_drop_extent_cache; } - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); if (disk_num_bytes < cur_alloc_size) break; @@ -1082,10 +1031,10 @@ out: return ret; out_drop_extent_cache: - btrfs_drop_extent_cache(inode, start, start + ram_size - 1, 0); + btrfs_drop_extent_cache(BTRFS_I(inode), start, start + ram_size - 1, 0); out_reserve: - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); - btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 1); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); out_unlock: extent_clear_unlock_delalloc(inode, start, end, delalloc_end, locked_page, @@ -1119,6 +1068,7 @@ static noinline void async_cow_start(struct btrfs_work *work) */ static noinline void async_cow_submit(struct btrfs_work *work) { + struct btrfs_fs_info *fs_info; struct async_cow *async_cow; struct btrfs_root *root; unsigned long nr_pages; @@ -1126,16 +1076,17 @@ static noinline void async_cow_submit(struct btrfs_work *work) async_cow = container_of(work, struct async_cow, work); root = async_cow->root; + fs_info = root->fs_info; nr_pages = (async_cow->end - async_cow->start + PAGE_SIZE) >> PAGE_SHIFT; /* * atomic_sub_return implies a barrier for waitqueue_active */ - if (atomic_sub_return(nr_pages, &root->fs_info->async_delalloc_pages) < + if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) < 5 * SZ_1M && - waitqueue_active(&root->fs_info->async_submit_wait)) - wake_up(&root->fs_info->async_submit_wait); + waitqueue_active(&fs_info->async_submit_wait)) + wake_up(&fs_info->async_submit_wait); if (async_cow->inode) submit_compressed_extents(async_cow->inode, async_cow); @@ -1154,11 +1105,11 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct async_cow *async_cow; struct btrfs_root *root = BTRFS_I(inode)->root; unsigned long nr_pages; u64 cur_end; - int limit = 10 * SZ_1M; clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, EXTENT_LOCKED, 1, 0, NULL, GFP_NOFS); @@ -1171,7 +1122,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, async_cow->start = start; if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS && - !btrfs_test_opt(root->fs_info, FORCE_COMPRESS)) + !btrfs_test_opt(fs_info, FORCE_COMPRESS)) cur_end = end; else cur_end = min(end, start + SZ_512K - 1); @@ -1186,22 +1137,15 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, nr_pages = (cur_end - start + PAGE_SIZE) >> PAGE_SHIFT; - atomic_add(nr_pages, &root->fs_info->async_delalloc_pages); - - btrfs_queue_work(root->fs_info->delalloc_workers, - &async_cow->work); + atomic_add(nr_pages, &fs_info->async_delalloc_pages); - if (atomic_read(&root->fs_info->async_delalloc_pages) > limit) { - wait_event(root->fs_info->async_submit_wait, - (atomic_read(&root->fs_info->async_delalloc_pages) < - limit)); - } + btrfs_queue_work(fs_info->delalloc_workers, &async_cow->work); - while (atomic_read(&root->fs_info->async_submit_draining) && - atomic_read(&root->fs_info->async_delalloc_pages)) { - wait_event(root->fs_info->async_submit_wait, - (atomic_read(&root->fs_info->async_delalloc_pages) == - 0)); + while (atomic_read(&fs_info->async_submit_draining) && + atomic_read(&fs_info->async_delalloc_pages)) { + wait_event(fs_info->async_submit_wait, + (atomic_read(&fs_info->async_delalloc_pages) == + 0)); } *nr_written += nr_pages; @@ -1211,14 +1155,14 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, return 0; } -static noinline int csum_exist_in_range(struct btrfs_root *root, +static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes) { int ret; struct btrfs_ordered_sum *sums; LIST_HEAD(list); - ret = btrfs_lookup_csums_range(root->fs_info->csum_root, bytenr, + ret = btrfs_lookup_csums_range(fs_info->csum_root, bytenr, bytenr + num_bytes - 1, &list, 0); if (ret == 0 && list_empty(&list)) return 0; @@ -1243,12 +1187,13 @@ static noinline int run_delalloc_nocow(struct inode *inode, u64 start, u64 end, int *page_started, int force, unsigned long *nr_written) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_trans_handle *trans; struct extent_buffer *leaf; struct btrfs_path *path; struct btrfs_file_extent_item *fi; struct btrfs_key found_key; + struct extent_map *em; u64 cow_start; u64 cur_offset; u64 extent_end; @@ -1263,7 +1208,7 @@ static noinline int run_delalloc_nocow(struct inode *inode, int nocow; int check_prev = 1; bool nolock; - u64 ino = btrfs_ino(inode); + u64 ino = btrfs_ino(BTRFS_I(inode)); path = btrfs_alloc_path(); if (!path) { @@ -1278,32 +1223,12 @@ static noinline int run_delalloc_nocow(struct inode *inode, return -ENOMEM; } - nolock = btrfs_is_free_space_inode(inode); - - if (nolock) - trans = btrfs_join_transaction_nolock(root); - else - trans = btrfs_join_transaction(root); - - if (IS_ERR(trans)) { - extent_clear_unlock_delalloc(inode, start, end, end, - locked_page, - EXTENT_LOCKED | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING | - EXTENT_DEFRAG, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | - PAGE_END_WRITEBACK); - btrfs_free_path(path); - return PTR_ERR(trans); - } - - trans->block_rsv = &root->fs_info->delalloc_block_rsv; + nolock = btrfs_is_free_space_inode(BTRFS_I(inode)); cow_start = (u64)-1; cur_offset = start; while (1) { - ret = btrfs_lookup_file_extent(trans, root, path, ino, + ret = btrfs_lookup_file_extent(NULL, root, path, ino, cur_offset, 0); if (ret < 0) goto error; @@ -1374,9 +1299,9 @@ next_slot: goto out_check; if (extent_type == BTRFS_FILE_EXTENT_REG && !force) goto out_check; - if (btrfs_extent_readonly(root, disk_bytenr)) + if (btrfs_extent_readonly(fs_info, disk_bytenr)) goto out_check; - if (btrfs_cross_ref_exist(trans, root, ino, + if (btrfs_cross_ref_exist(root, ino, found_key.offset - extent_offset, disk_bytenr)) goto out_check; @@ -1397,17 +1322,24 @@ next_slot: * this ensure that csum for a given extent are * either valid or do not exist. */ - if (csum_exist_in_range(root, disk_bytenr, num_bytes)) + if (csum_exist_in_range(fs_info, disk_bytenr, + num_bytes)) { + if (!nolock) + btrfs_end_write_no_snapshoting(root); goto out_check; - if (!btrfs_inc_nocow_writers(root->fs_info, - disk_bytenr)) + } + if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) { + if (!nolock) + btrfs_end_write_no_snapshoting(root); goto out_check; + } nocow = 1; } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { extent_end = found_key.offset + btrfs_file_extent_inline_len(leaf, path->slots[0], fi); - extent_end = ALIGN(extent_end, root->sectorsize); + extent_end = ALIGN(extent_end, + fs_info->sectorsize); } else { BUG_ON(1); } @@ -1417,8 +1349,7 @@ out_check: if (!nolock && nocow) btrfs_end_write_no_snapshoting(root); if (nocow) - btrfs_dec_nocow_writers(root->fs_info, - disk_bytenr); + btrfs_dec_nocow_writers(fs_info, disk_bytenr); goto next_slot; } if (!nocow) { @@ -1441,7 +1372,7 @@ out_check: if (!nolock && nocow) btrfs_end_write_no_snapshoting(root); if (nocow) - btrfs_dec_nocow_writers(root->fs_info, + btrfs_dec_nocow_writers(fs_info, disk_bytenr); goto error; } @@ -1449,35 +1380,28 @@ out_check: } if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) { - struct extent_map *em; - struct extent_map_tree *em_tree; - em_tree = &BTRFS_I(inode)->extent_tree; - em = alloc_extent_map(); - BUG_ON(!em); /* -ENOMEM */ - em->start = cur_offset; - em->orig_start = found_key.offset - extent_offset; - em->len = num_bytes; - em->block_len = num_bytes; - em->block_start = disk_bytenr; - em->orig_block_len = disk_num_bytes; - em->ram_bytes = ram_bytes; - em->bdev = 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, 1); - write_unlock(&em_tree->lock); - if (ret != -EEXIST) { - free_extent_map(em); - break; - } - btrfs_drop_extent_cache(inode, em->start, - em->start + em->len - 1, 0); + u64 orig_start = found_key.offset - extent_offset; + + em = create_io_em(inode, cur_offset, num_bytes, + orig_start, + disk_bytenr, /* block_start */ + num_bytes, /* block_len */ + disk_num_bytes, /* orig_block_len */ + ram_bytes, BTRFS_COMPRESS_NONE, + BTRFS_ORDERED_PREALLOC); + if (IS_ERR(em)) { + if (!nolock && nocow) + btrfs_end_write_no_snapshoting(root); + if (nocow) + btrfs_dec_nocow_writers(fs_info, + disk_bytenr); + ret = PTR_ERR(em); + goto error; } + free_extent_map(em); + } + + if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) { type = BTRFS_ORDERED_PREALLOC; } else { type = BTRFS_ORDERED_NOCOW; @@ -1486,7 +1410,7 @@ out_check: ret = btrfs_add_ordered_extent(inode, cur_offset, disk_bytenr, num_bytes, num_bytes, type); if (nocow) - btrfs_dec_nocow_writers(root->fs_info, disk_bytenr); + btrfs_dec_nocow_writers(fs_info, disk_bytenr); BUG_ON(ret); /* -ENOMEM */ if (root->root_key.objectid == @@ -1528,10 +1452,6 @@ out_check: } error: - err = btrfs_end_transaction(trans, root); - if (!ret) - ret = err; - if (ret && cur_offset < end) extent_clear_unlock_delalloc(inode, cur_offset, end, end, locked_page, EXTENT_LOCKED | @@ -1603,7 +1523,7 @@ static void btrfs_split_extent_hook(struct inode *inode, size = orig->end - orig->start + 1; if (size > BTRFS_MAX_EXTENT_SIZE) { - u64 num_extents; + u32 num_extents; u64 new_size; /* @@ -1611,13 +1531,10 @@ static void btrfs_split_extent_hook(struct inode *inode, * applies here, just in reverse. */ new_size = orig->end - split + 1; - num_extents = div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); + num_extents = count_max_extents(new_size); new_size = split - orig->start; - num_extents += div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); - if (div64_u64(size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE) >= num_extents) + num_extents += count_max_extents(new_size); + if (count_max_extents(size) >= num_extents) return; } @@ -1637,7 +1554,7 @@ static void btrfs_merge_extent_hook(struct inode *inode, struct extent_state *other) { u64 new_size, old_size; - u64 num_extents; + u32 num_extents; /* not delalloc, ignore it */ if (!(other->state & EXTENT_DELALLOC)) @@ -1675,14 +1592,10 @@ static void btrfs_merge_extent_hook(struct inode *inode, * this case. */ old_size = other->end - other->start + 1; - num_extents = div64_u64(old_size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); + num_extents = count_max_extents(old_size); old_size = new->end - new->start + 1; - num_extents += div64_u64(old_size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); - - if (div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE) >= num_extents) + num_extents += count_max_extents(old_size); + if (count_max_extents(new_size) >= num_extents) return; spin_lock(&BTRFS_I(inode)->lock); @@ -1693,6 +1606,8 @@ static void btrfs_merge_extent_hook(struct inode *inode, static void btrfs_add_delalloc_inodes(struct btrfs_root *root, struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + spin_lock(&root->delalloc_lock); if (list_empty(&BTRFS_I(inode)->delalloc_inodes)) { list_add_tail(&BTRFS_I(inode)->delalloc_inodes, @@ -1701,30 +1616,32 @@ static void btrfs_add_delalloc_inodes(struct btrfs_root *root, &BTRFS_I(inode)->runtime_flags); root->nr_delalloc_inodes++; if (root->nr_delalloc_inodes == 1) { - spin_lock(&root->fs_info->delalloc_root_lock); + spin_lock(&fs_info->delalloc_root_lock); BUG_ON(!list_empty(&root->delalloc_root)); list_add_tail(&root->delalloc_root, - &root->fs_info->delalloc_roots); - spin_unlock(&root->fs_info->delalloc_root_lock); + &fs_info->delalloc_roots); + spin_unlock(&fs_info->delalloc_root_lock); } } spin_unlock(&root->delalloc_lock); } static void btrfs_del_delalloc_inode(struct btrfs_root *root, - struct inode *inode) + struct btrfs_inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); + spin_lock(&root->delalloc_lock); - if (!list_empty(&BTRFS_I(inode)->delalloc_inodes)) { - list_del_init(&BTRFS_I(inode)->delalloc_inodes); + if (!list_empty(&inode->delalloc_inodes)) { + list_del_init(&inode->delalloc_inodes); clear_bit(BTRFS_INODE_IN_DELALLOC_LIST, - &BTRFS_I(inode)->runtime_flags); + &inode->runtime_flags); root->nr_delalloc_inodes--; if (!root->nr_delalloc_inodes) { - spin_lock(&root->fs_info->delalloc_root_lock); + spin_lock(&fs_info->delalloc_root_lock); BUG_ON(list_empty(&root->delalloc_root)); list_del_init(&root->delalloc_root); - spin_unlock(&root->fs_info->delalloc_root_lock); + spin_unlock(&fs_info->delalloc_root_lock); } } spin_unlock(&root->delalloc_lock); @@ -1739,6 +1656,8 @@ static void btrfs_set_bit_hook(struct inode *inode, struct extent_state *state, unsigned *bits) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + if ((*bits & EXTENT_DEFRAG) && !(*bits & EXTENT_DELALLOC)) WARN_ON(1); /* @@ -1749,7 +1668,7 @@ static void btrfs_set_bit_hook(struct inode *inode, if (!(state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; - bool do_list = !btrfs_is_free_space_inode(inode); + bool do_list = !btrfs_is_free_space_inode(BTRFS_I(inode)); if (*bits & EXTENT_FIRST_DELALLOC) { *bits &= ~EXTENT_FIRST_DELALLOC; @@ -1760,11 +1679,11 @@ static void btrfs_set_bit_hook(struct inode *inode, } /* For sanity tests */ - if (btrfs_is_testing(root->fs_info)) + if (btrfs_is_testing(fs_info)) return; - __percpu_counter_add(&root->fs_info->delalloc_bytes, len, - root->fs_info->delalloc_batch); + __percpu_counter_add(&fs_info->delalloc_bytes, len, + fs_info->delalloc_batch); spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->delalloc_bytes += len; if (*bits & EXTENT_DEFRAG) @@ -1779,18 +1698,18 @@ 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, +static void btrfs_clear_bit_hook(struct btrfs_inode *inode, struct extent_state *state, unsigned *bits) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); u64 len = state->end + 1 - state->start; - u64 num_extents = div64_u64(len + BTRFS_MAX_EXTENT_SIZE -1, - BTRFS_MAX_EXTENT_SIZE); + u32 num_extents = count_max_extents(len); - spin_lock(&BTRFS_I(inode)->lock); + spin_lock(&inode->lock); if ((state->state & EXTENT_DEFRAG) && (*bits & EXTENT_DEFRAG)) - BTRFS_I(inode)->defrag_bytes -= len; - spin_unlock(&BTRFS_I(inode)->lock); + inode->defrag_bytes -= len; + spin_unlock(&inode->lock); /* * set_bit and clear bit hooks normally require _irqsave/restore @@ -1798,15 +1717,15 @@ static void btrfs_clear_bit_hook(struct inode *inode, * bit, which is only set or cleared with irqs on */ if ((state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; bool do_list = !btrfs_is_free_space_inode(inode); if (*bits & EXTENT_FIRST_DELALLOC) { *bits &= ~EXTENT_FIRST_DELALLOC; } else if (!(*bits & EXTENT_DO_ACCOUNTING)) { - spin_lock(&BTRFS_I(inode)->lock); - BTRFS_I(inode)->outstanding_extents -= num_extents; - spin_unlock(&BTRFS_I(inode)->lock); + spin_lock(&inode->lock); + inode->outstanding_extents -= num_extents; + spin_unlock(&inode->lock); } /* @@ -1815,29 +1734,30 @@ static void btrfs_clear_bit_hook(struct inode *inode, * error. */ if (*bits & EXTENT_DO_ACCOUNTING && - root != root->fs_info->tree_root) + root != fs_info->tree_root) btrfs_delalloc_release_metadata(inode, len); /* For sanity tests. */ - if (btrfs_is_testing(root->fs_info)) + if (btrfs_is_testing(fs_info)) return; if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID && do_list && !(state->state & EXTENT_NORESERVE) && (*bits & (EXTENT_DO_ACCOUNTING | EXTENT_CLEAR_DATA_RESV))) - btrfs_free_reserved_data_space_noquota(inode, + btrfs_free_reserved_data_space_noquota( + &inode->vfs_inode, state->start, 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 && + __percpu_counter_add(&fs_info->delalloc_bytes, -len, + fs_info->delalloc_batch); + spin_lock(&inode->lock); + inode->delalloc_bytes -= len; + if (do_list && inode->delalloc_bytes == 0 && test_bit(BTRFS_INODE_IN_DELALLOC_LIST, - &BTRFS_I(inode)->runtime_flags)) + &inode->runtime_flags)) btrfs_del_delalloc_inode(root, inode); - spin_unlock(&BTRFS_I(inode)->lock); + spin_unlock(&inode->lock); } } @@ -1853,7 +1773,8 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset, size_t size, struct bio *bio, unsigned long bio_flags) { - struct btrfs_root *root = BTRFS_I(page->mapping->host)->root; + struct inode *inode = page->mapping->host; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 logical = (u64)bio->bi_iter.bi_sector << 9; u64 length = 0; u64 map_length; @@ -1864,8 +1785,8 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset, length = bio->bi_iter.bi_size; map_length = length; - ret = btrfs_map_block(root->fs_info, bio_op(bio), logical, - &map_length, NULL, 0); + ret = btrfs_map_block(fs_info, btrfs_op(bio), logical, &map_length, + NULL, 0); if (ret < 0) return ret; if (map_length < length + size) @@ -1885,10 +1806,9 @@ static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { - struct btrfs_root *root = BTRFS_I(inode)->root; int ret = 0; - ret = btrfs_csum_one_bio(root, inode, bio, 0, 0); + ret = btrfs_csum_one_bio(inode, bio, 0, 0); BUG_ON(ret); /* -ENOMEM */ return 0; } @@ -1905,10 +1825,10 @@ static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int ret; - ret = btrfs_map_bio(root, bio, mirror_num, 1); + ret = btrfs_map_bio(fs_info, bio, mirror_num, 1); if (ret) { bio->bi_error = ret; bio_endio(bio); @@ -1924,6 +1844,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; enum btrfs_wq_endio_type metadata = BTRFS_WQ_ENDIO_DATA; int ret = 0; @@ -1932,11 +1853,11 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - if (btrfs_is_free_space_inode(inode)) + if (btrfs_is_free_space_inode(BTRFS_I(inode))) metadata = BTRFS_WQ_ENDIO_FREE_SPACE; if (bio_op(bio) != REQ_OP_WRITE) { - ret = btrfs_bio_wq_end_io(root->fs_info, bio, metadata); + ret = btrfs_bio_wq_end_io(fs_info, bio, metadata); if (ret) goto out; @@ -1946,7 +1867,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, bio_flags); goto out; } else if (!skip_sum) { - ret = btrfs_lookup_bio_sums(root, inode, bio, NULL); + ret = btrfs_lookup_bio_sums(inode, bio, NULL); if (ret) goto out; } @@ -1956,20 +1877,19 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID) goto mapit; /* we're doing a write, do the async checksumming */ - ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info, - inode, bio, mirror_num, - bio_flags, bio_offset, - __btrfs_submit_bio_start, - __btrfs_submit_bio_done); + ret = btrfs_wq_submit_bio(fs_info, inode, bio, mirror_num, + bio_flags, bio_offset, + __btrfs_submit_bio_start, + __btrfs_submit_bio_done); goto out; } else if (!skip_sum) { - ret = btrfs_csum_one_bio(root, inode, bio, 0, 0); + ret = btrfs_csum_one_bio(inode, bio, 0, 0); if (ret) goto out; } mapit: - ret = btrfs_map_bio(root, bio, mirror_num, 0); + ret = btrfs_map_bio(fs_info, bio, mirror_num, 0); out: if (ret < 0) { @@ -1984,8 +1904,7 @@ out: * at IO completion time based on sums calculated at bio submission time. */ static noinline int add_pending_csums(struct btrfs_trans_handle *trans, - struct inode *inode, u64 file_offset, - struct list_head *list) + struct inode *inode, struct list_head *list) { struct btrfs_ordered_sum *sum; @@ -2043,7 +1962,7 @@ again: if (PagePrivate2(page)) goto out; - ordered = btrfs_lookup_ordered_range(inode, page_start, + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start, PAGE_SIZE); if (ordered) { unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, @@ -2090,8 +2009,8 @@ out_page: static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end) { struct inode *inode = page->mapping->host; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_writepage_fixup *fixup; - struct btrfs_root *root = BTRFS_I(inode)->root; /* this page is properly in the ordered list */ if (TestClearPagePrivate2(page)) @@ -2109,7 +2028,7 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end) btrfs_init_work(&fixup->work, btrfs_fixup_helper, btrfs_writepage_fixup_worker, NULL, NULL); fixup->page = page; - btrfs_queue_work(root->fs_info->fixup_workers, &fixup->work); + btrfs_queue_work(fs_info->fixup_workers, &fixup->work); return -EBUSY; } @@ -2148,7 +2067,7 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, goto out; if (!extent_inserted) { - ins.objectid = btrfs_ino(inode); + ins.objectid = btrfs_ino(BTRFS_I(inode)); ins.offset = file_pos; ins.type = BTRFS_EXTENT_DATA_KEY; @@ -2180,10 +2099,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, ins.objectid = disk_bytenr; ins.offset = disk_num_bytes; ins.type = BTRFS_EXTENT_ITEM_KEY; - ret = btrfs_alloc_reserved_file_extent(trans, root, - root->root_key.objectid, - btrfs_ino(inode), file_pos, - ram_bytes, &ins); + ret = btrfs_alloc_reserved_file_extent(trans, root->root_key.objectid, + btrfs_ino(BTRFS_I(inode)), file_pos, ram_bytes, &ins); /* * Release the reserved range from inode dirty range map, as it is * already moved into delayed_ref_head @@ -2293,7 +2210,6 @@ 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; @@ -2302,20 +2218,20 @@ static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id, struct sa_defrag_extent_backref *backref; struct extent_buffer *leaf; struct inode *inode = new->inode; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int slot; int ret; u64 extent_offset; u64 num_bytes; if (BTRFS_I(inode)->root->root_key.objectid == root_id && - inum == btrfs_ino(inode)) + inum == btrfs_ino(BTRFS_I(inode))) return 0; key.objectid = root_id; 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) @@ -2413,7 +2329,7 @@ out: 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 btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb); struct old_sa_defrag_extent *old, *tmp; int ret; @@ -2471,13 +2387,12 @@ static noinline int relink_extent_backref(struct btrfs_path *path, 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 btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb); struct inode *inode; struct extent_state *cached = NULL; int ret = 0; @@ -2498,7 +2413,6 @@ static noinline int relink_extent_backref(struct btrfs_path *path, 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); @@ -2580,7 +2494,7 @@ static noinline int relink_extent_backref(struct btrfs_path *path, if (ret) goto out_free_path; again: - key.objectid = btrfs_ino(inode); + key.objectid = btrfs_ino(BTRFS_I(inode)); key.type = BTRFS_EXTENT_DATA_KEY; key.offset = start; @@ -2643,7 +2557,7 @@ again: inode_add_bytes(inode, len); btrfs_release_path(path); - ret = btrfs_inc_extent_ref(trans, root, new->bytenr, + ret = btrfs_inc_extent_ref(trans, fs_info, new->bytenr, new->disk_len, 0, backref->root_id, backref->inum, new->file_pos); /* start - extent_offset */ @@ -2656,7 +2570,7 @@ again: out_free_path: btrfs_release_path(path); path->leave_spinning = 0; - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); out_unlock: unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, lock_end, &cached, GFP_NOFS); @@ -2679,6 +2593,7 @@ static void free_sa_defrag_extent(struct new_sa_defrag_extent *new) static void relink_file_extents(struct new_sa_defrag_extent *new) { + struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb); struct btrfs_path *path; struct sa_defrag_extent_backref *backref; struct sa_defrag_extent_backref *prev = NULL; @@ -2725,14 +2640,15 @@ static void relink_file_extents(struct new_sa_defrag_extent *new) out: free_sa_defrag_extent(new); - atomic_dec(&root->fs_info->defrag_running); - wake_up(&root->fs_info->transaction_wait); + atomic_dec(&fs_info->defrag_running); + wake_up(&fs_info->transaction_wait); } static struct new_sa_defrag_extent * record_old_file_extents(struct inode *inode, struct btrfs_ordered_extent *ordered) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_path *path; struct btrfs_key key; @@ -2757,7 +2673,7 @@ record_old_file_extents(struct inode *inode, if (!path) goto out_kfree; - key.objectid = btrfs_ino(inode); + key.objectid = btrfs_ino(BTRFS_I(inode)); key.type = BTRFS_EXTENT_DATA_KEY; key.offset = new->file_pos; @@ -2792,7 +2708,7 @@ record_old_file_extents(struct inode *inode, btrfs_item_key_to_cpu(l, &key, slot); - if (key.objectid != btrfs_ino(inode)) + if (key.objectid != btrfs_ino(BTRFS_I(inode))) break; if (key.type != BTRFS_EXTENT_DATA_KEY) break; @@ -2831,7 +2747,7 @@ next: } btrfs_free_path(path); - atomic_inc(&root->fs_info->defrag_running); + atomic_inc(&fs_info->defrag_running); return new; @@ -2842,12 +2758,12 @@ out_kfree: return NULL; } -static void btrfs_release_delalloc_bytes(struct btrfs_root *root, +static void btrfs_release_delalloc_bytes(struct btrfs_fs_info *fs_info, u64 start, u64 len) { struct btrfs_block_group_cache *cache; - cache = btrfs_lookup_block_group(root->fs_info, start); + cache = btrfs_lookup_block_group(fs_info, start); ASSERT(cache); spin_lock(&cache->lock); @@ -2864,6 +2780,7 @@ static void btrfs_release_delalloc_bytes(struct btrfs_root *root, static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) { struct inode *inode = ordered_extent->inode; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans = NULL; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; @@ -2875,16 +2792,17 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) bool nolock; bool truncated = false; - nolock = btrfs_is_free_space_inode(inode); + nolock = btrfs_is_free_space_inode(BTRFS_I(inode)); if (test_bit(BTRFS_ORDERED_IOERR, &ordered_extent->flags)) { ret = -EIO; goto out; } - btrfs_free_io_failure_record(inode, ordered_extent->file_offset, - ordered_extent->file_offset + - ordered_extent->len - 1); + btrfs_free_io_failure_record(BTRFS_I(inode), + ordered_extent->file_offset, + ordered_extent->file_offset + + ordered_extent->len - 1); if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) { truncated = true; @@ -2914,7 +2832,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) trans = NULL; goto out; } - trans->block_rsv = &root->fs_info->delalloc_block_rsv; + trans->block_rsv = &fs_info->delalloc_block_rsv; ret = btrfs_update_inode_fallback(trans, root, inode); if (ret) /* -ENOMEM or corruption */ btrfs_abort_transaction(trans, ret); @@ -2949,18 +2867,18 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out_unlock; } - trans->block_rsv = &root->fs_info->delalloc_block_rsv; + trans->block_rsv = &fs_info->delalloc_block_rsv; if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags)) compress_type = ordered_extent->compress_type; if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) { BUG_ON(compress_type); - ret = btrfs_mark_extent_written(trans, inode, + ret = btrfs_mark_extent_written(trans, BTRFS_I(inode), ordered_extent->file_offset, ordered_extent->file_offset + logical_len); } else { - BUG_ON(root == root->fs_info->tree_root); + BUG_ON(root == fs_info->tree_root); ret = insert_reserved_file_extent(trans, inode, ordered_extent->file_offset, ordered_extent->start, @@ -2969,7 +2887,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) compress_type, 0, 0, BTRFS_FILE_EXTENT_REG); if (!ret) - btrfs_release_delalloc_bytes(root, + btrfs_release_delalloc_bytes(fs_info, ordered_extent->start, ordered_extent->disk_len); } @@ -2981,8 +2899,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out_unlock; } - add_pending_csums(trans, inode, ordered_extent->file_offset, - &ordered_extent->list); + add_pending_csums(trans, inode, &ordered_extent->list); btrfs_ordered_update_i_size(inode, 0, ordered_extent); ret = btrfs_update_inode_fallback(trans, root, inode); @@ -2996,10 +2913,11 @@ out_unlock: ordered_extent->file_offset + ordered_extent->len - 1, &cached_state, GFP_NOFS); out: - if (root != root->fs_info->tree_root) - btrfs_delalloc_release_metadata(inode, ordered_extent->len); + if (root != fs_info->tree_root) + btrfs_delalloc_release_metadata(BTRFS_I(inode), + ordered_extent->len); if (trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (ret || truncated) { u64 start, end; @@ -3012,7 +2930,7 @@ out: clear_extent_uptodate(io_tree, start, end, NULL, GFP_NOFS); /* Drop the cache for the part of the extent we didn't write. */ - btrfs_drop_extent_cache(inode, start, end, 0); + btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0); /* * If the ordered extent had an IOERR or something else went @@ -3023,7 +2941,8 @@ out: if ((ret || !logical_len) && !test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) && !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) - btrfs_free_reserved_extent(root, ordered_extent->start, + btrfs_free_reserved_extent(fs_info, + ordered_extent->start, ordered_extent->disk_len, 1); } @@ -3038,7 +2957,7 @@ out: if (new) { if (ret) { free_sa_defrag_extent(new); - atomic_dec(&root->fs_info->defrag_running); + atomic_dec(&fs_info->defrag_running); } else { relink_file_extents(new); } @@ -3059,11 +2978,11 @@ static void finish_ordered_fn(struct btrfs_work *work) btrfs_finish_ordered_io(ordered_extent); } -static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end, +static void btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end, struct extent_state *state, int uptodate) { struct inode *inode = page->mapping->host; - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_ordered_extent *ordered_extent = NULL; struct btrfs_workqueue *wq; btrfs_work_func_t func; @@ -3073,21 +2992,19 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end, ClearPagePrivate2(page); if (!btrfs_dec_test_ordered_pending(inode, &ordered_extent, start, end - start + 1, uptodate)) - return 0; + return; - if (btrfs_is_free_space_inode(inode)) { - wq = root->fs_info->endio_freespace_worker; + if (btrfs_is_free_space_inode(BTRFS_I(inode))) { + wq = fs_info->endio_freespace_worker; func = btrfs_freespace_write_helper; } else { - wq = root->fs_info->endio_write_workers; + wq = fs_info->endio_write_workers; func = btrfs_endio_write_helper; } btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL, NULL); btrfs_queue_work(wq, &ordered_extent->work); - - return 0; } static int __readpage_endio_check(struct inode *inode, @@ -3103,16 +3020,15 @@ static int __readpage_endio_check(struct inode *inode, kaddr = kmap_atomic(page); csum = btrfs_csum_data(kaddr + pgoff, csum, len); - btrfs_csum_final(csum, (char *)&csum); + btrfs_csum_final(csum, (u8 *)&csum); if (csum != csum_expected) goto zeroit; kunmap_atomic(kaddr); return 0; zeroit: - btrfs_warn_rl(BTRFS_I(inode)->root->fs_info, - "csum failed ino %llu off %llu csum %u expected csum %u", - btrfs_ino(inode), start, csum, csum_expected); + btrfs_print_data_csum_error(BTRFS_I(inode), start, csum, csum_expected, + io_bio->mirror_num); memset(kaddr + pgoff, 1, len); flush_dcache_page(page); kunmap_atomic(kaddr); @@ -3156,7 +3072,7 @@ static int btrfs_readpage_end_io_hook(struct btrfs_io_bio *io_bio, void btrfs_add_delayed_iput(struct inode *inode) { - struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_inode *binode = BTRFS_I(inode); if (atomic_add_unless(&inode->i_count, -1, 1)) @@ -3172,9 +3088,8 @@ void btrfs_add_delayed_iput(struct inode *inode) spin_unlock(&fs_info->delayed_iput_lock); } -void btrfs_run_delayed_iputs(struct btrfs_root *root) +void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info) { - struct btrfs_fs_info *fs_info = root->fs_info; spin_lock(&fs_info->delayed_iput_lock); while (!list_empty(&fs_info->delayed_iputs)) { @@ -3204,6 +3119,7 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root) void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, struct btrfs_root *root) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_block_rsv *block_rsv; int ret; @@ -3228,7 +3144,7 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state) && btrfs_root_refs(&root->root_item) > 0) { - ret = btrfs_del_orphan_item(trans, root->fs_info->tree_root, + ret = btrfs_del_orphan_item(trans, fs_info->tree_root, root->root_key.objectid); if (ret) btrfs_abort_transaction(trans, ret); @@ -3239,7 +3155,7 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, if (block_rsv) { WARN_ON(block_rsv->size > 0); - btrfs_free_block_rsv(root, block_rsv); + btrfs_free_block_rsv(fs_info, block_rsv); } } @@ -3250,16 +3166,19 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, * NOTE: caller of this function should reserve 5 units of metadata for * this function. */ -int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) +int btrfs_orphan_add(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); + struct btrfs_root *root = inode->root; struct btrfs_block_rsv *block_rsv = NULL; int reserve = 0; int insert = 0; int ret; if (!root->orphan_block_rsv) { - block_rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); + block_rsv = btrfs_alloc_block_rsv(fs_info, + BTRFS_BLOCK_RSV_TEMP); if (!block_rsv) return -ENOMEM; } @@ -3268,12 +3187,12 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) if (!root->orphan_block_rsv) { root->orphan_block_rsv = block_rsv; } else if (block_rsv) { - btrfs_free_block_rsv(root, block_rsv); + btrfs_free_block_rsv(fs_info, block_rsv); block_rsv = NULL; } if (!test_and_set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags)) { + &inode->runtime_flags)) { #if 0 /* * For proper ENOSPC handling, we should do orphan @@ -3290,7 +3209,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) } if (!test_and_set_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &BTRFS_I(inode)->runtime_flags)) + &inode->runtime_flags)) reserve = 1; spin_unlock(&root->orphan_lock); @@ -3301,10 +3220,10 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) if (ret) { atomic_dec(&root->orphan_inodes); clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &BTRFS_I(inode)->runtime_flags); + &inode->runtime_flags); if (insert) clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags); + &inode->runtime_flags); return ret; } } @@ -3316,12 +3235,12 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) atomic_dec(&root->orphan_inodes); if (reserve) { clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &BTRFS_I(inode)->runtime_flags); + &inode->runtime_flags); btrfs_orphan_release_metadata(inode); } if (ret != -EEXIST) { clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags); + &inode->runtime_flags); btrfs_abort_transaction(trans, ret); return ret; } @@ -3331,7 +3250,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) /* insert an orphan item to track subvolume contains orphan files */ if (insert >= 2) { - ret = btrfs_insert_orphan_item(trans, root->fs_info->tree_root, + ret = btrfs_insert_orphan_item(trans, fs_info->tree_root, root->root_key.objectid); if (ret && ret != -EEXIST) { btrfs_abort_transaction(trans, ret); @@ -3346,20 +3265,20 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) * item for this particular inode. */ static int btrfs_orphan_del(struct btrfs_trans_handle *trans, - struct inode *inode) + struct btrfs_inode *inode) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; int delete_item = 0; int release_rsv = 0; int ret = 0; spin_lock(&root->orphan_lock); if (test_and_clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags)) + &inode->runtime_flags)) delete_item = 1; if (test_and_clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &BTRFS_I(inode)->runtime_flags)) + &inode->runtime_flags)) release_rsv = 1; spin_unlock(&root->orphan_lock); @@ -3382,6 +3301,7 @@ static int btrfs_orphan_del(struct btrfs_trans_handle *trans, */ int btrfs_orphan_cleanup(struct btrfs_root *root) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_key key, found_key; @@ -3441,8 +3361,8 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) */ if (found_key.offset == last_objectid) { - btrfs_err(root->fs_info, - "Error removing orphan entry, stopping orphan cleanup"); + btrfs_err(fs_info, + "Error removing orphan entry, stopping orphan cleanup"); ret = -EINVAL; goto out; } @@ -3452,12 +3372,12 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) 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); + inode = btrfs_iget(fs_info->sb, &found_key, root, NULL); ret = PTR_ERR_OR_ZERO(inode); if (ret && ret != -ENOENT) goto out; - if (ret == -ENOENT && root == root->fs_info->tree_root) { + if (ret == -ENOENT && root == fs_info->tree_root) { struct btrfs_root *dead_root; struct btrfs_fs_info *fs_info = root->fs_info; int is_dead_root = 0; @@ -3499,11 +3419,11 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) ret = PTR_ERR(trans); goto out; } - btrfs_debug(root->fs_info, "auto deleting %Lu", - found_key.objectid); + btrfs_debug(fs_info, "auto deleting %Lu", + found_key.objectid); ret = btrfs_del_orphan_item(trans, root, found_key.objectid); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (ret) goto out; continue; @@ -3532,8 +3452,8 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) ret = PTR_ERR(trans); goto out; } - ret = btrfs_orphan_add(trans, inode); - btrfs_end_transaction(trans, root); + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); + btrfs_end_transaction(trans); if (ret) { iput(inode); goto out; @@ -3541,7 +3461,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) ret = btrfs_truncate(inode); if (ret) - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); } else { nr_unlink++; } @@ -3557,25 +3477,24 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE; if (root->orphan_block_rsv) - btrfs_block_rsv_release(root, root->orphan_block_rsv, + btrfs_block_rsv_release(fs_info, root->orphan_block_rsv, (u64)-1); if (root->orphan_block_rsv || test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) { trans = btrfs_join_transaction(root); if (!IS_ERR(trans)) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); } if (nr_unlink) - btrfs_debug(root->fs_info, "unlinked %d orphans", nr_unlink); + btrfs_debug(fs_info, "unlinked %d orphans", nr_unlink); if (nr_truncate) - btrfs_debug(root->fs_info, "truncated %d orphans", nr_truncate); + btrfs_debug(fs_info, "truncated %d orphans", nr_truncate); out: if (ret) - btrfs_err(root->fs_info, - "could not do orphan cleanup %d", ret); + btrfs_err(fs_info, "could not do orphan cleanup %d", ret); btrfs_free_path(path); return ret; } @@ -3654,6 +3573,7 @@ static noinline int acls_after_inode_item(struct extent_buffer *leaf, */ static int btrfs_read_locked_inode(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_inode_item *inode_item; @@ -3696,7 +3616,7 @@ static int btrfs_read_locked_inode(struct inode *inode) set_nlink(inode, btrfs_inode_nlink(leaf, inode_item)); i_uid_write(inode, btrfs_inode_uid(leaf, inode_item)); i_gid_write(inode, btrfs_inode_gid(leaf, inode_item)); - btrfs_i_size_write(inode, btrfs_inode_size(leaf, inode_item)); + btrfs_i_size_write(BTRFS_I(inode), btrfs_inode_size(leaf, inode_item)); inode->i_atime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->atime); inode->i_atime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->atime); @@ -3734,7 +3654,7 @@ cache_index: * This is required for both inode re-read from disk and delayed inode * in delayed_nodes_tree. */ - if (BTRFS_I(inode)->last_trans == root->fs_info->generation) + if (BTRFS_I(inode)->last_trans == fs_info->generation) set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); @@ -3773,7 +3693,7 @@ cache_index: goto cache_acl; btrfs_item_key_to_cpu(leaf, &location, path->slots[0]); - if (location.objectid != btrfs_ino(inode)) + if (location.objectid != btrfs_ino(BTRFS_I(inode))) goto cache_acl; ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); @@ -3795,14 +3715,14 @@ cache_acl: * any xattrs or acls */ maybe_acls = acls_after_inode_item(leaf, path->slots[0], - btrfs_ino(inode), &first_xattr_slot); + btrfs_ino(BTRFS_I(inode)), &first_xattr_slot); if (first_xattr_slot != -1) { path->slots[0] = first_xattr_slot; ret = btrfs_load_inode_props(inode, path); if (ret) - btrfs_err(root->fs_info, + btrfs_err(fs_info, "error loading props for ino %llu (root %llu): %d", - btrfs_ino(inode), + btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, ret); } btrfs_free_path(path); @@ -3819,10 +3739,7 @@ cache_acl: break; case S_IFDIR: inode->i_fop = &btrfs_dir_file_operations; - if (root == root->fs_info->tree_root) - inode->i_op = &btrfs_dir_ro_inode_operations; - else - inode->i_op = &btrfs_dir_inode_operations; + inode->i_op = &btrfs_dir_inode_operations; break; case S_IFLNK: inode->i_op = &btrfs_symlink_inode_operations; @@ -3937,6 +3854,7 @@ failed: noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode) { + struct btrfs_fs_info *fs_info = root->fs_info; int ret; /* @@ -3946,9 +3864,9 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, * The data relocation inode should also be directly updated * without delay */ - if (!btrfs_is_free_space_inode(inode) + if (!btrfs_is_free_space_inode(BTRFS_I(inode)) && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID - && !test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) { + && !test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) { btrfs_update_root_times(trans, root); ret = btrfs_delayed_update_inode(trans, root, inode); @@ -3979,9 +3897,11 @@ noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, */ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, - struct inode *dir, struct inode *inode, + struct btrfs_inode *dir, + struct btrfs_inode *inode, const char *name, int name_len) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; int ret = 0; struct extent_buffer *leaf; @@ -4025,10 +3945,10 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, * that we delay to delete it, and just do this deletion when * we update the inode item. */ - if (BTRFS_I(inode)->dir_index) { + if (inode->dir_index) { ret = btrfs_delayed_delete_inode_ref(inode); if (!ret) { - index = BTRFS_I(inode)->dir_index; + index = inode->dir_index; goto skip_backref; } } @@ -4036,28 +3956,28 @@ 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) { - btrfs_info(root->fs_info, + btrfs_info(fs_info, "failed to delete reference to %.*s, inode %llu parent %llu", name_len, name, ino, dir_ino); btrfs_abort_transaction(trans, ret); goto err; } skip_backref: - ret = btrfs_delete_delayed_dir_index(trans, root, dir, index); + ret = btrfs_delete_delayed_dir_index(trans, fs_info, dir, index); if (ret) { btrfs_abort_transaction(trans, ret); goto err; } - ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len, - inode, dir_ino); + ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len, inode, + dir_ino); if (ret != 0 && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); goto err; } - ret = btrfs_del_dir_entries_in_log(trans, root, name, name_len, - dir, index); + ret = btrfs_del_dir_entries_in_log(trans, root, name, name_len, dir, + index); if (ret == -ENOENT) ret = 0; else if (ret) @@ -4067,26 +3987,26 @@ err: if (ret) goto out; - btrfs_i_size_write(dir, dir->i_size - name_len * 2); - inode_inc_iversion(inode); - inode_inc_iversion(dir); - inode->i_ctime = dir->i_mtime = - dir->i_ctime = current_time(inode); - ret = btrfs_update_inode(trans, root, dir); + btrfs_i_size_write(dir, dir->vfs_inode.i_size - name_len * 2); + inode_inc_iversion(&inode->vfs_inode); + inode_inc_iversion(&dir->vfs_inode); + inode->vfs_inode.i_ctime = dir->vfs_inode.i_mtime = + dir->vfs_inode.i_ctime = current_time(&inode->vfs_inode); + ret = btrfs_update_inode(trans, root, &dir->vfs_inode); out: return ret; } int btrfs_unlink_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, - struct inode *dir, struct inode *inode, + struct btrfs_inode *dir, struct btrfs_inode *inode, const char *name, int name_len) { int ret; ret = __btrfs_unlink_inode(trans, root, dir, inode, name, name_len); if (!ret) { - drop_nlink(inode); - ret = btrfs_update_inode(trans, root, inode); + drop_nlink(&inode->vfs_inode); + ret = btrfs_update_inode(trans, root, &inode->vfs_inode); } return ret; } @@ -4124,22 +4044,24 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry) if (IS_ERR(trans)) return PTR_ERR(trans); - btrfs_record_unlink_dir(trans, dir, d_inode(dentry), 0); + btrfs_record_unlink_dir(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), + 0); - ret = btrfs_unlink_inode(trans, root, dir, d_inode(dentry), - dentry->d_name.name, dentry->d_name.len); + ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir), + BTRFS_I(d_inode(dentry)), dentry->d_name.name, + dentry->d_name.len); if (ret) goto out; if (inode->i_nlink == 0) { - ret = btrfs_orphan_add(trans, inode); + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); if (ret) goto out; } out: - btrfs_end_transaction(trans, root); - btrfs_btree_balance_dirty(root); + btrfs_end_transaction(trans); + btrfs_btree_balance_dirty(root->fs_info); return ret; } @@ -4148,13 +4070,14 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, struct inode *dir, u64 objectid, const char *name, int name_len) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_dir_item *di; struct btrfs_key key; u64 index; int ret; - u64 dir_ino = btrfs_ino(dir); + u64 dir_ino = btrfs_ino(BTRFS_I(dir)); path = btrfs_alloc_path(); if (!path) @@ -4180,9 +4103,9 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, } btrfs_release_path(path); - ret = btrfs_del_root_ref(trans, root->fs_info->tree_root, - objectid, root->root_key.objectid, - dir_ino, &index, name, name_len); + ret = btrfs_del_root_ref(trans, fs_info, objectid, + root->root_key.objectid, dir_ino, + &index, name, name_len); if (ret < 0) { if (ret != -ENOENT) { btrfs_abort_transaction(trans, ret); @@ -4206,13 +4129,13 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, } btrfs_release_path(path); - ret = btrfs_delete_delayed_dir_index(trans, root, dir, index); + ret = btrfs_delete_delayed_dir_index(trans, fs_info, BTRFS_I(dir), index); if (ret) { btrfs_abort_transaction(trans, ret); goto out; } - btrfs_i_size_write(dir, dir->i_size - name_len * 2); + btrfs_i_size_write(BTRFS_I(dir), dir->i_size - name_len * 2); inode_inc_iversion(dir); dir->i_mtime = dir->i_ctime = current_time(dir); ret = btrfs_update_inode_fallback(trans, root, dir); @@ -4233,14 +4156,14 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) return -ENOTEMPTY; - if (btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID) + if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID) return -EPERM; trans = __unlink_start_trans(dir); if (IS_ERR(trans)) return PTR_ERR(trans); - if (unlikely(btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { + if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { err = btrfs_unlink_subvol(trans, root, dir, BTRFS_I(inode)->location.objectid, dentry->d_name.name, @@ -4248,17 +4171,18 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) goto out; } - err = btrfs_orphan_add(trans, inode); + err = btrfs_orphan_add(trans, BTRFS_I(inode)); if (err) goto out; last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; /* now the directory is empty */ - err = btrfs_unlink_inode(trans, root, dir, d_inode(dentry), - dentry->d_name.name, dentry->d_name.len); + err = btrfs_unlink_inode(trans, root, BTRFS_I(dir), + BTRFS_I(d_inode(dentry)), dentry->d_name.name, + dentry->d_name.len); if (!err) { - btrfs_i_size_write(inode, 0); + btrfs_i_size_write(BTRFS_I(inode), 0); /* * Propagate the last_unlink_trans value of the deleted dir to * its parent directory. This is to prevent an unrecoverable @@ -4274,8 +4198,8 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) BTRFS_I(dir)->last_unlink_trans = last_unlink_trans; } out: - btrfs_end_transaction(trans, root); - btrfs_btree_balance_dirty(root); + btrfs_end_transaction(trans); + btrfs_btree_balance_dirty(root->fs_info); return err; } @@ -4284,18 +4208,19 @@ static int truncate_space_check(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytes_deleted) { + struct btrfs_fs_info *fs_info = root->fs_info; int ret; /* * This is only used to apply pressure to the enospc system, we don't * intend to use this reservation at all. */ - bytes_deleted = btrfs_csum_bytes_to_leaves(root, bytes_deleted); - bytes_deleted *= root->nodesize; - ret = btrfs_block_rsv_add(root, &root->fs_info->trans_block_rsv, + bytes_deleted = btrfs_csum_bytes_to_leaves(fs_info, bytes_deleted); + bytes_deleted *= fs_info->nodesize; + ret = btrfs_block_rsv_add(root, &fs_info->trans_block_rsv, bytes_deleted, BTRFS_RESERVE_NO_FLUSH); if (!ret) { - trace_btrfs_space_reservation(root->fs_info, "transaction", + trace_btrfs_space_reservation(fs_info, "transaction", trans->transid, bytes_deleted, 1); trans->bytes_reserved += bytes_deleted; @@ -4338,7 +4263,7 @@ static int truncate_inline_extent(struct inode *inode, btrfs_set_file_extent_ram_bytes(leaf, fi, size); size = btrfs_file_extent_calc_inline_size(size); - btrfs_truncate_item(root, path, size, 1); + btrfs_truncate_item(root->fs_info, path, size, 1); if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) inode_sub_bytes(inode, item_end + 1 - new_size); @@ -4362,6 +4287,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, struct inode *inode, u64 new_size, u32 min_type) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; @@ -4380,7 +4306,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, int extent_type = -1; int ret; int err = 0; - u64 ino = btrfs_ino(inode); + u64 ino = btrfs_ino(BTRFS_I(inode)); u64 bytes_deleted = 0; bool be_nice = 0; bool should_throttle = 0; @@ -4392,7 +4318,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, * for non-free space inodes and ref cows, we want to back off from * time to time */ - if (!btrfs_is_free_space_inode(inode) && + if (!btrfs_is_free_space_inode(BTRFS_I(inode)) && test_bit(BTRFS_ROOT_REF_COWS, &root->state)) be_nice = 1; @@ -4407,9 +4333,10 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, * extent just the way it is. */ if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) || - root == root->fs_info->tree_root) - btrfs_drop_extent_cache(inode, ALIGN(new_size, - root->sectorsize), (u64)-1, 0); + root == fs_info->tree_root) + btrfs_drop_extent_cache(BTRFS_I(inode), ALIGN(new_size, + fs_info->sectorsize), + (u64)-1, 0); /* * This function is also used to drop the items in the log tree before @@ -4418,7 +4345,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, * items. */ if (min_type == 0 && root == BTRFS_I(inode)->root) - btrfs_kill_delayed_inode_items(inode); + btrfs_kill_delayed_inode_items(BTRFS_I(inode)); key.objectid = ino; key.offset = (u64)-1; @@ -4431,7 +4358,7 @@ search_again: * bytes_deleted is > 0, it will be huge by the time we get here */ if (be_nice && bytes_deleted > SZ_32M) { - if (btrfs_should_end_transaction(trans, root)) { + if (btrfs_should_end_transaction(trans)) { err = -EAGAIN; goto error; } @@ -4508,7 +4435,7 @@ search_again: btrfs_file_extent_num_bytes(leaf, fi); extent_num_bytes = ALIGN(new_size - found_key.offset, - root->sectorsize); + fs_info->sectorsize); btrfs_set_file_extent_num_bytes(leaf, fi, extent_num_bytes); num_dec = (orig_num_bytes - @@ -4595,16 +4522,16 @@ delete: if (found_extent && (test_bit(BTRFS_ROOT_REF_COWS, &root->state) || - root == root->fs_info->tree_root)) { + root == fs_info->tree_root)) { btrfs_set_path_blocking(path); bytes_deleted += extent_num_bytes; - ret = btrfs_free_extent(trans, root, extent_start, + ret = btrfs_free_extent(trans, fs_info, extent_start, extent_num_bytes, 0, btrfs_header_owner(leaf), ino, extent_offset); BUG_ON(ret); - if (btrfs_should_throttle_delayed_refs(trans, root)) - btrfs_async_run_delayed_refs(root, + if (btrfs_should_throttle_delayed_refs(trans, fs_info)) + btrfs_async_run_delayed_refs(fs_info, trans->delayed_ref_updates * 2, trans->transid, 0); if (be_nice) { @@ -4613,9 +4540,8 @@ delete: should_end = 1; } if (btrfs_should_throttle_delayed_refs(trans, - root)) { + fs_info)) should_throttle = 1; - } } } @@ -4640,7 +4566,9 @@ delete: unsigned long updates = trans->delayed_ref_updates; if (updates) { trans->delayed_ref_updates = 0; - ret = btrfs_run_delayed_refs(trans, root, updates * 2); + ret = btrfs_run_delayed_refs(trans, + fs_info, + updates * 2); if (ret && !err) err = ret; } @@ -4666,16 +4594,28 @@ out: btrfs_abort_transaction(trans, ret); } error: - if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) + if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { + ASSERT(last_size >= new_size); + if (!err && last_size > new_size) + last_size = new_size; btrfs_ordered_update_i_size(inode, last_size, NULL); + } btrfs_free_path(path); + if (err == 0) { + /* only inline file may have last_size != new_size */ + if (new_size >= fs_info->sectorsize || + new_size > fs_info->max_inline) + ASSERT(last_size == new_size); + } + if (be_nice && bytes_deleted > SZ_32M) { unsigned long updates = trans->delayed_ref_updates; if (updates) { trans->delayed_ref_updates = 0; - ret = btrfs_run_delayed_refs(trans, root, updates * 2); + ret = btrfs_run_delayed_refs(trans, fs_info, + updates * 2); if (ret && !err) err = ret; } @@ -4697,13 +4637,13 @@ error: int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, int front) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct address_space *mapping = inode->i_mapping; - struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; char *kaddr; - u32 blocksize = root->sectorsize; + u32 blocksize = fs_info->sectorsize; pgoff_t index = from >> PAGE_SHIFT; unsigned offset = from & (blocksize - 1); struct page *page; @@ -4807,6 +4747,7 @@ out: static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, u64 offset, u64 len) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_trans_handle *trans; int ret; @@ -4814,8 +4755,8 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, * Still need to make sure the inode looks like it's been updated so * that any holes get logged if we fsync. */ - if (btrfs_fs_incompat(root->fs_info, NO_HOLES)) { - BTRFS_I(inode)->last_trans = root->fs_info->generation; + if (btrfs_fs_incompat(fs_info, NO_HOLES)) { + BTRFS_I(inode)->last_trans = fs_info->generation; BTRFS_I(inode)->last_sub_trans = root->log_transid; BTRFS_I(inode)->last_log_commit = root->last_log_commit; return 0; @@ -4833,17 +4774,17 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, ret = btrfs_drop_extents(trans, root, inode, offset, offset + len, 1); if (ret) { btrfs_abort_transaction(trans, ret); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); return ret; } - ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset, - 0, 0, len, 0, len, 0, 0, 0); + ret = btrfs_insert_file_extent(trans, root, btrfs_ino(BTRFS_I(inode)), + offset, 0, 0, len, 0, len, 0, 0, 0); if (ret) btrfs_abort_transaction(trans, ret); else btrfs_update_inode(trans, root, inode); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); return ret; } @@ -4855,13 +4796,14 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, */ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct extent_map *em = NULL; struct extent_state *cached_state = NULL; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; - u64 hole_start = ALIGN(oldsize, root->sectorsize); - u64 block_end = ALIGN(size, root->sectorsize); + u64 hole_start = ALIGN(oldsize, fs_info->sectorsize); + u64 block_end = ALIGN(size, fs_info->sectorsize); u64 last_byte; u64 cur_offset; u64 hole_size; @@ -4884,7 +4826,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) lock_extent_bits(io_tree, hole_start, block_end - 1, &cached_state); - ordered = btrfs_lookup_ordered_range(inode, hole_start, + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), hole_start, block_end - hole_start); if (!ordered) break; @@ -4896,7 +4838,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) cur_offset = hole_start; while (1) { - em = btrfs_get_extent(inode, NULL, 0, cur_offset, + em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset, block_end - cur_offset, 0); if (IS_ERR(em)) { err = PTR_ERR(em); @@ -4904,7 +4846,7 @@ 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 = ALIGN(last_byte , root->sectorsize); + last_byte = ALIGN(last_byte, fs_info->sectorsize); if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { struct extent_map *hole_em; hole_size = last_byte - cur_offset; @@ -4913,7 +4855,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) hole_size); if (err) break; - btrfs_drop_extent_cache(inode, cur_offset, + btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, cur_offset + hole_size - 1, 0); hole_em = alloc_extent_map(); if (!hole_em) { @@ -4929,9 +4871,9 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) 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->bdev = fs_info->fs_devices->latest_bdev; hole_em->compress_type = BTRFS_COMPRESS_NONE; - hole_em->generation = root->fs_info->generation; + hole_em->generation = fs_info->generation; while (1) { write_lock(&em_tree->lock); @@ -4939,7 +4881,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) write_unlock(&em_tree->lock); if (err != -EEXIST) break; - btrfs_drop_extent_cache(inode, cur_offset, + btrfs_drop_extent_cache(BTRFS_I(inode), + cur_offset, cur_offset + hole_size - 1, 0); } @@ -5006,7 +4949,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) pagecache_isize_extended(inode, oldsize, newsize); ret = btrfs_update_inode(trans, root, inode); btrfs_end_write_no_snapshoting(root); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); } else { /* @@ -5036,8 +4979,8 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) * so we need to guarantee from this point on that everything * will be consistent. */ - ret = btrfs_orphan_add(trans, inode); - btrfs_end_transaction(trans, root); + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); + btrfs_end_transaction(trans); if (ret) return ret; @@ -5045,14 +4988,21 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) truncate_setsize(inode, newsize); /* Disable nonlocked read DIO to avoid the end less truncate */ - btrfs_inode_block_unlocked_dio(inode); + btrfs_inode_block_unlocked_dio(BTRFS_I(inode)); inode_dio_wait(inode); - btrfs_inode_resume_unlocked_dio(inode); + btrfs_inode_resume_unlocked_dio(BTRFS_I(inode)); ret = btrfs_truncate(inode); if (ret && inode->i_nlink) { int err; + /* To get a stable disk_i_size */ + err = btrfs_wait_ordered_range(inode, 0, (u64)-1); + if (err) { + btrfs_orphan_del(NULL, BTRFS_I(inode)); + return err; + } + /* * failed to truncate, disk_i_size is only adjusted down * as we remove extents, so it should represent the true @@ -5061,14 +5011,14 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) */ trans = btrfs_join_transaction(root); if (IS_ERR(trans)) { - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); return ret; } i_size_write(inode, BTRFS_I(inode)->disk_i_size); - err = btrfs_orphan_del(trans, inode); + err = btrfs_orphan_del(trans, BTRFS_I(inode)); if (err) btrfs_abort_transaction(trans, err); - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); } } @@ -5201,6 +5151,7 @@ static void evict_inode_truncate_pages(struct inode *inode) void btrfs_evict_inode(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *rsv, *global_rsv; @@ -5215,27 +5166,27 @@ void btrfs_evict_inode(struct inode *inode) return; } - min_size = btrfs_calc_trunc_metadata_size(root, 1); + min_size = btrfs_calc_trunc_metadata_size(fs_info, 1); evict_inode_truncate_pages(inode); if (inode->i_nlink && ((btrfs_root_refs(&root->root_item) != 0 && root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID) || - btrfs_is_free_space_inode(inode))) + btrfs_is_free_space_inode(BTRFS_I(inode)))) goto no_delete; if (is_bad_inode(inode)) { - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); goto no_delete; } /* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */ if (!special_file(inode->i_mode)) btrfs_wait_ordered_range(inode, 0, (u64)-1); - btrfs_free_io_failure_record(inode, 0, (u64)-1); + btrfs_free_io_failure_record(BTRFS_I(inode), 0, (u64)-1); - if (test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) { + if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) { BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, &BTRFS_I(inode)->runtime_flags)); goto no_delete; @@ -5247,22 +5198,22 @@ void btrfs_evict_inode(struct inode *inode) goto no_delete; } - ret = btrfs_commit_inode_delayed_inode(inode); + ret = btrfs_commit_inode_delayed_inode(BTRFS_I(inode)); if (ret) { - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); goto no_delete; } - rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); + rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); if (!rsv) { - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); goto no_delete; } rsv->size = min_size; rsv->failfast = 1; - global_rsv = &root->fs_info->global_block_rsv; + global_rsv = &fs_info->global_block_rsv; - btrfs_i_size_write(inode, 0); + btrfs_i_size_write(BTRFS_I(inode), 0); /* * This is a bit simpler than btrfs_truncate since we've already @@ -5294,18 +5245,18 @@ void btrfs_evict_inode(struct inode *inode) * steal_from_global == 3: abandon all hope! */ if (steal_from_global > 2) { - 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); + btrfs_warn(fs_info, + "Could not get space for a delete, will truncate on mount %d", + ret); + btrfs_orphan_del(NULL, BTRFS_I(inode)); + btrfs_free_block_rsv(fs_info, rsv); goto no_delete; } trans = btrfs_join_transaction(root); if (IS_ERR(trans)) { - btrfs_orphan_del(NULL, inode); - btrfs_free_block_rsv(root, rsv); + btrfs_orphan_del(NULL, BTRFS_I(inode)); + btrfs_free_block_rsv(fs_info, rsv); goto no_delete; } @@ -5315,7 +5266,7 @@ void btrfs_evict_inode(struct inode *inode) * again. */ if (steal_from_global) { - if (!btrfs_check_space_for_delayed_refs(trans, root)) + if (!btrfs_check_space_for_delayed_refs(trans, fs_info)) ret = btrfs_block_rsv_migrate(global_rsv, rsv, min_size, 0); else @@ -5328,10 +5279,10 @@ void btrfs_evict_inode(struct inode *inode) * again. */ if (ret) { - ret = btrfs_commit_transaction(trans, root); + ret = btrfs_commit_transaction(trans); if (ret) { - btrfs_orphan_del(NULL, inode); - btrfs_free_block_rsv(root, rsv); + btrfs_orphan_del(NULL, BTRFS_I(inode)); + btrfs_free_block_rsv(fs_info, rsv); goto no_delete; } continue; @@ -5345,13 +5296,13 @@ void btrfs_evict_inode(struct inode *inode) if (ret != -ENOSPC && ret != -EAGAIN) break; - trans->block_rsv = &root->fs_info->trans_block_rsv; - btrfs_end_transaction(trans, root); + trans->block_rsv = &fs_info->trans_block_rsv; + btrfs_end_transaction(trans); trans = NULL; - btrfs_btree_balance_dirty(root); + btrfs_btree_balance_dirty(fs_info); } - btrfs_free_block_rsv(root, rsv); + btrfs_free_block_rsv(fs_info, rsv); /* * Errors here aren't a big deal, it just means we leave orphan items @@ -5359,20 +5310,20 @@ void btrfs_evict_inode(struct inode *inode) */ if (ret == 0) { trans->block_rsv = root->orphan_block_rsv; - btrfs_orphan_del(trans, inode); + btrfs_orphan_del(trans, BTRFS_I(inode)); } else { - btrfs_orphan_del(NULL, inode); + btrfs_orphan_del(NULL, BTRFS_I(inode)); } - trans->block_rsv = &root->fs_info->trans_block_rsv; - if (!(root == root->fs_info->tree_root || + trans->block_rsv = &fs_info->trans_block_rsv; + if (!(root == fs_info->tree_root || root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)) - btrfs_return_ino(root, btrfs_ino(inode)); + btrfs_return_ino(root, btrfs_ino(BTRFS_I(inode))); - btrfs_end_transaction(trans, root); - btrfs_btree_balance_dirty(root); + btrfs_end_transaction(trans); + btrfs_btree_balance_dirty(fs_info); no_delete: - btrfs_remove_delayed_node(inode); + btrfs_remove_delayed_node(BTRFS_I(inode)); clear_inode(inode); } @@ -5394,8 +5345,8 @@ static int btrfs_inode_by_name(struct inode *dir, struct dentry *dentry, if (!path) return -ENOMEM; - di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(dir), name, - namelen, 0); + di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(BTRFS_I(dir)), + name, namelen, 0); if (IS_ERR(di)) ret = PTR_ERR(di); @@ -5416,7 +5367,7 @@ out_err: * needs to be changed to reflect the root directory of the tree root. This * is kind of like crossing a mount point. */ -static int fixup_tree_root_location(struct btrfs_root *root, +static int fixup_tree_root_location(struct btrfs_fs_info *fs_info, struct inode *dir, struct dentry *dentry, struct btrfs_key *location, @@ -5441,8 +5392,7 @@ static int fixup_tree_root_location(struct btrfs_root *root, key.type = BTRFS_ROOT_REF_KEY; key.offset = location->objectid; - ret = btrfs_search_slot(NULL, root->fs_info->tree_root, &key, path, - 0, 0); + ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); if (ret) { if (ret < 0) err = ret; @@ -5451,7 +5401,7 @@ static int fixup_tree_root_location(struct btrfs_root *root, leaf = path->nodes[0]; ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_root_ref); - if (btrfs_root_ref_dirid(leaf, ref) != btrfs_ino(dir) || + if (btrfs_root_ref_dirid(leaf, ref) != btrfs_ino(BTRFS_I(dir)) || btrfs_root_ref_name_len(leaf, ref) != dentry->d_name.len) goto out; @@ -5463,7 +5413,7 @@ static int fixup_tree_root_location(struct btrfs_root *root, btrfs_release_path(path); - new_root = btrfs_read_fs_root_no_name(root->fs_info, location); + new_root = btrfs_read_fs_root_no_name(fs_info, location); if (IS_ERR(new_root)) { err = PTR_ERR(new_root); goto out; @@ -5486,7 +5436,7 @@ static void inode_tree_add(struct inode *inode) struct rb_node **p; struct rb_node *parent; struct rb_node *new = &BTRFS_I(inode)->rb_node; - u64 ino = btrfs_ino(inode); + u64 ino = btrfs_ino(BTRFS_I(inode)); if (inode_unhashed(inode)) return; @@ -5497,9 +5447,9 @@ static void inode_tree_add(struct inode *inode) parent = *p; entry = rb_entry(parent, struct btrfs_inode, rb_node); - if (ino < btrfs_ino(&entry->vfs_inode)) + if (ino < btrfs_ino(BTRFS_I(&entry->vfs_inode))) p = &parent->rb_left; - else if (ino > btrfs_ino(&entry->vfs_inode)) + else if (ino > btrfs_ino(BTRFS_I(&entry->vfs_inode))) p = &parent->rb_right; else { WARN_ON(!(entry->vfs_inode.i_state & @@ -5517,6 +5467,7 @@ static void inode_tree_add(struct inode *inode) static void inode_tree_del(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; int empty = 0; @@ -5529,7 +5480,7 @@ static void inode_tree_del(struct inode *inode) spin_unlock(&root->inode_lock); if (empty && btrfs_root_refs(&root->root_item) == 0) { - synchronize_srcu(&root->fs_info->subvol_srcu); + synchronize_srcu(&fs_info->subvol_srcu); spin_lock(&root->inode_lock); empty = RB_EMPTY_ROOT(&root->inode_tree); spin_unlock(&root->inode_lock); @@ -5540,13 +5491,14 @@ static void inode_tree_del(struct inode *inode) void btrfs_invalidate_inodes(struct btrfs_root *root) { + struct btrfs_fs_info *fs_info = root->fs_info; struct rb_node *node; struct rb_node *prev; struct btrfs_inode *entry; struct inode *inode; u64 objectid = 0; - if (!test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state)) + if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) WARN_ON(btrfs_root_refs(&root->root_item) != 0); spin_lock(&root->inode_lock); @@ -5557,9 +5509,9 @@ again: prev = node; entry = rb_entry(node, struct btrfs_inode, rb_node); - if (objectid < btrfs_ino(&entry->vfs_inode)) + if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode))) node = node->rb_left; - else if (objectid > btrfs_ino(&entry->vfs_inode)) + else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode))) node = node->rb_right; else break; @@ -5567,7 +5519,7 @@ again: if (!node) { while (prev) { entry = rb_entry(prev, struct btrfs_inode, rb_node); - if (objectid <= btrfs_ino(&entry->vfs_inode)) { + if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) { node = prev; break; } @@ -5576,7 +5528,7 @@ again: } while (node) { entry = rb_entry(node, struct btrfs_inode, rb_node); - objectid = btrfs_ino(&entry->vfs_inode) + 1; + objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1; inode = igrab(&entry->vfs_inode); if (inode) { spin_unlock(&root->inode_lock); @@ -5682,6 +5634,7 @@ static struct inode *new_simple_dir(struct super_block *s, inode->i_ino = BTRFS_EMPTY_SUBVOL_DIR_OBJECTID; inode->i_op = &btrfs_dir_ro_inode_operations; + inode->i_opflags &= ~IOP_XATTR; inode->i_fop = &simple_dir_operations; inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO; inode->i_mtime = current_time(inode); @@ -5694,6 +5647,7 @@ static struct inode *new_simple_dir(struct super_block *s, struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct inode *inode; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_root *sub_root = root; @@ -5718,8 +5672,8 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) BUG_ON(location.type != BTRFS_ROOT_ITEM_KEY); - index = srcu_read_lock(&root->fs_info->subvol_srcu); - ret = fixup_tree_root_location(root, dir, dentry, + index = srcu_read_lock(&fs_info->subvol_srcu); + ret = fixup_tree_root_location(fs_info, dir, dentry, &location, &sub_root); if (ret < 0) { if (ret != -ENOENT) @@ -5729,13 +5683,13 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) } else { inode = btrfs_iget(dir->i_sb, &location, sub_root, NULL); } - srcu_read_unlock(&root->fs_info->subvol_srcu, index); + srcu_read_unlock(&fs_info->subvol_srcu, index); if (!IS_ERR(inode) && root != sub_root) { - down_read(&root->fs_info->cleanup_work_sem); + down_read(&fs_info->cleanup_work_sem); if (!(inode->i_sb->s_flags & MS_RDONLY)) ret = btrfs_orphan_cleanup(sub_root); - up_read(&root->fs_info->cleanup_work_sem); + up_read(&fs_info->cleanup_work_sem); if (ret) { iput(inode); inode = ERR_PTR(ret); @@ -5758,7 +5712,7 @@ static int btrfs_dentry_delete(const struct dentry *dentry) if (btrfs_root_refs(&root->root_item) == 0) return 1; - if (btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) + if (btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) return 1; } return 0; @@ -5792,6 +5746,7 @@ unsigned char btrfs_filetype_table[] = { static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_item *item; struct btrfs_dir_item *di; @@ -5805,20 +5760,11 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) int slot; unsigned char d_type; int over = 0; - u32 di_cur; - u32 di_total; - u32 di_len; - int key_type = BTRFS_DIR_INDEX_KEY; char tmp_name[32]; char *name_ptr; int name_len; - int is_curr = 0; /* ctx->pos points to the current index? */ - bool emitted; bool put = false; - - /* FIXME, use a real flag for deciding about the key type */ - if (root->fs_info->tree_root == root) - key_type = BTRFS_DIR_ITEM_KEY; + struct btrfs_key location; if (!dir_emit_dots(file, ctx)) return 0; @@ -5829,22 +5775,18 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) path->reada = READA_FORWARD; - if (key_type == BTRFS_DIR_INDEX_KEY) { - INIT_LIST_HEAD(&ins_list); - INIT_LIST_HEAD(&del_list); - put = btrfs_readdir_get_delayed_items(inode, &ins_list, - &del_list); - } + INIT_LIST_HEAD(&ins_list); + INIT_LIST_HEAD(&del_list); + put = btrfs_readdir_get_delayed_items(inode, &ins_list, &del_list); - key.type = key_type; + key.type = BTRFS_DIR_INDEX_KEY; key.offset = ctx->pos; - key.objectid = btrfs_ino(inode); + key.objectid = btrfs_ino(BTRFS_I(inode)); ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) goto err; - emitted = false; while (1) { leaf = path->nodes[0]; slot = path->slots[0]; @@ -5862,98 +5804,52 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) if (found_key.objectid != key.objectid) break; - if (found_key.type != key_type) + if (found_key.type != BTRFS_DIR_INDEX_KEY) break; if (found_key.offset < ctx->pos) goto next; - if (key_type == BTRFS_DIR_INDEX_KEY && - btrfs_should_delete_dir_index(&del_list, - found_key.offset)) + if (btrfs_should_delete_dir_index(&del_list, found_key.offset)) goto next; ctx->pos = found_key.offset; - is_curr = 1; di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item); - di_cur = 0; - di_total = btrfs_item_size(leaf, item); - - while (di_cur < di_total) { - struct btrfs_key location; - - if (verify_dir_item(root, leaf, di)) - break; + if (verify_dir_item(fs_info, leaf, di)) + goto next; - name_len = btrfs_dir_name_len(leaf, di); - if (name_len <= sizeof(tmp_name)) { - name_ptr = tmp_name; - } else { - name_ptr = kmalloc(name_len, GFP_KERNEL); - if (!name_ptr) { - ret = -ENOMEM; - goto err; - } + name_len = btrfs_dir_name_len(leaf, di); + if (name_len <= sizeof(tmp_name)) { + name_ptr = tmp_name; + } else { + name_ptr = kmalloc(name_len, GFP_KERNEL); + if (!name_ptr) { + ret = -ENOMEM; + goto err; } - read_extent_buffer(leaf, name_ptr, - (unsigned long)(di + 1), name_len); - - d_type = btrfs_filetype_table[btrfs_dir_type(leaf, di)]; - btrfs_dir_item_key_to_cpu(leaf, di, &location); + } + read_extent_buffer(leaf, name_ptr, (unsigned long)(di + 1), + name_len); + d_type = btrfs_filetype_table[btrfs_dir_type(leaf, di)]; + btrfs_dir_item_key_to_cpu(leaf, di, &location); - /* is this a reference to our own snapshot? If so - * skip it. - * - * In contrast to old kernels, we insert the snapshot's - * dir item and dir index after it has been created, so - * we won't find a reference to our own snapshot. We - * still keep the following code for backward - * compatibility. - */ - if (location.type == BTRFS_ROOT_ITEM_KEY && - location.objectid == root->root_key.objectid) { - over = 0; - goto skip; - } - over = !dir_emit(ctx, name_ptr, name_len, - location.objectid, d_type); + over = !dir_emit(ctx, name_ptr, name_len, location.objectid, + d_type); -skip: - if (name_ptr != tmp_name) - kfree(name_ptr); + if (name_ptr != tmp_name) + kfree(name_ptr); - if (over) - goto nopos; - emitted = true; - di_len = btrfs_dir_name_len(leaf, di) + - btrfs_dir_data_len(leaf, di) + sizeof(*di); - di_cur += di_len; - di = (struct btrfs_dir_item *)((char *)di + di_len); - } + if (over) + goto nopos; + ctx->pos++; next: path->slots[0]++; } - if (key_type == BTRFS_DIR_INDEX_KEY) { - if (is_curr) - ctx->pos++; - ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list, &emitted); - if (ret) - goto nopos; - } - - /* - * If we haven't emitted any dir entry, we must not touch ctx->pos as - * it was was set to the termination value in previous call. We assume - * that "." and ".." were emitted if we reach this point and set the - * termination value as well for an empty directory. - */ - if (ctx->pos > 2 && !emitted) + ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list); + if (ret) goto nopos; - /* Reached end of directory/root. Bump pos past the last item. */ - ctx->pos++; - /* * Stop new entries from being returned after we return the last * entry. @@ -5971,12 +5867,10 @@ next: * last entry requires it because doing so has broken 32bit apps * in the past. */ - if (key_type == BTRFS_DIR_INDEX_KEY) { - if (ctx->pos >= INT_MAX) - ctx->pos = LLONG_MAX; - else - ctx->pos = INT_MAX; - } + if (ctx->pos >= INT_MAX) + ctx->pos = LLONG_MAX; + else + ctx->pos = INT_MAX; nopos: ret = 0; err: @@ -5996,7 +5890,8 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) if (test_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags)) return 0; - if (btrfs_fs_closing(root->fs_info) && btrfs_is_free_space_inode(inode)) + if (btrfs_fs_closing(root->fs_info) && + btrfs_is_free_space_inode(BTRFS_I(inode))) nolock = true; if (wbc->sync_mode == WB_SYNC_ALL) { @@ -6006,7 +5901,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) trans = btrfs_join_transaction(root); if (IS_ERR(trans)) return PTR_ERR(trans); - ret = btrfs_commit_transaction(trans, root); + ret = btrfs_commit_transaction(trans); } return ret; } @@ -6019,6 +5914,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) */ static int btrfs_dirty_inode(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; int ret; @@ -6033,16 +5929,16 @@ static int btrfs_dirty_inode(struct inode *inode) ret = btrfs_update_inode(trans, root, inode); if (ret && ret == -ENOSPC) { /* whoops, lets try again with the full transaction */ - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, inode); } - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (BTRFS_I(inode)->delayed_node) - btrfs_balance_delayed_items(root); + btrfs_balance_delayed_items(fs_info); return ret; } @@ -6075,9 +5971,9 @@ static int btrfs_update_time(struct inode *inode, struct timespec *now, * and then set the in-memory index_cnt variable to reflect * free sequence numbers */ -static int btrfs_set_inode_index_count(struct inode *inode) +static int btrfs_set_inode_index_count(struct btrfs_inode *inode) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; struct btrfs_key key, found_key; struct btrfs_path *path; struct extent_buffer *leaf; @@ -6106,7 +6002,7 @@ static int btrfs_set_inode_index_count(struct inode *inode) * else has to start at 2 */ if (path->slots[0] == 0) { - BTRFS_I(inode)->index_cnt = 2; + inode->index_cnt = 2; goto out; } @@ -6117,11 +6013,11 @@ static int btrfs_set_inode_index_count(struct inode *inode) if (found_key.objectid != btrfs_ino(inode) || found_key.type != BTRFS_DIR_INDEX_KEY) { - BTRFS_I(inode)->index_cnt = 2; + inode->index_cnt = 2; goto out; } - BTRFS_I(inode)->index_cnt = found_key.offset + 1; + inode->index_cnt = found_key.offset + 1; out: btrfs_free_path(path); return ret; @@ -6131,11 +6027,11 @@ out: * helper to find a free sequence number in a given directory. This current * code is very simple, later versions will do smarter things in the btree */ -int btrfs_set_inode_index(struct inode *dir, u64 *index) +int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index) { int ret = 0; - if (BTRFS_I(dir)->index_cnt == (u64)-1) { + if (dir->index_cnt == (u64)-1) { ret = btrfs_inode_delayed_dir_index_count(dir); if (ret) { ret = btrfs_set_inode_index_count(dir); @@ -6144,8 +6040,8 @@ int btrfs_set_inode_index(struct inode *dir, u64 *index) } } - *index = BTRFS_I(dir)->index_cnt; - BTRFS_I(dir)->index_cnt++; + *index = dir->index_cnt; + dir->index_cnt++; return ret; } @@ -6168,6 +6064,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, u64 ref_objectid, u64 objectid, umode_t mode, u64 *index) { + struct btrfs_fs_info *fs_info = root->fs_info; struct inode *inode; struct btrfs_inode_item *inode_item; struct btrfs_key *location; @@ -6183,7 +6080,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, if (!path) return ERR_PTR(-ENOMEM); - inode = new_inode(root->fs_info->sb); + inode = new_inode(fs_info->sb); if (!inode) { btrfs_free_path(path); return ERR_PTR(-ENOMEM); @@ -6205,7 +6102,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, if (dir && name) { trace_btrfs_inode_request(dir); - ret = btrfs_set_inode_index(dir, index); + ret = btrfs_set_inode_index(BTRFS_I(dir), index); if (ret) { btrfs_free_path(path); iput(inode); @@ -6277,7 +6174,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_inode_item); - memset_extent_buffer(path->nodes[0], 0, (unsigned long)inode_item, + memzero_extent_buffer(path->nodes[0], (unsigned long)inode_item, sizeof(*inode_item)); fill_inode_item(trans, path->nodes[0], inode_item, inode); @@ -6296,9 +6193,9 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, btrfs_inherit_iflags(inode, dir); if (S_ISREG(mode)) { - if (btrfs_test_opt(root->fs_info, NODATASUM)) + if (btrfs_test_opt(fs_info, NODATASUM)) BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM; - if (btrfs_test_opt(root->fs_info, NODATACOW)) + if (btrfs_test_opt(fs_info, NODATACOW)) BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW | BTRFS_INODE_NODATASUM; } @@ -6312,9 +6209,9 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, ret = btrfs_inode_inherit_props(trans, inode, dir); if (ret) - btrfs_err(root->fs_info, + btrfs_err(fs_info, "error inheriting props for ino %llu (root %llu): %d", - btrfs_ino(inode), root->root_key.objectid, ret); + btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, ret); return inode; @@ -6340,17 +6237,18 @@ static inline u8 btrfs_inode_type(struct inode *inode) * inode to the parent directory. */ int btrfs_add_link(struct btrfs_trans_handle *trans, - struct inode *parent_inode, struct inode *inode, + struct btrfs_inode *parent_inode, struct btrfs_inode *inode, const char *name, int name_len, int add_backref, u64 index) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); int ret = 0; struct btrfs_key key; - struct btrfs_root *root = BTRFS_I(parent_inode)->root; + struct btrfs_root *root = parent_inode->root; u64 ino = btrfs_ino(inode); u64 parent_ino = btrfs_ino(parent_inode); if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { - memcpy(&key, &BTRFS_I(inode)->root->root_key, sizeof(key)); + memcpy(&key, &inode->root->root_key, sizeof(key)); } else { key.objectid = ino; key.type = BTRFS_INODE_ITEM_KEY; @@ -6358,9 +6256,9 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, } if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { - ret = btrfs_add_root_ref(trans, root->fs_info->tree_root, - key.objectid, root->root_key.objectid, - parent_ino, index, name, name_len); + ret = btrfs_add_root_ref(trans, fs_info, key.objectid, + root->root_key.objectid, parent_ino, + index, name, name_len); } else if (add_backref) { ret = btrfs_insert_inode_ref(trans, root, name, name_len, ino, parent_ino, index); @@ -6372,7 +6270,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, ret = btrfs_insert_dir_item(trans, root, name, name_len, parent_inode, &key, - btrfs_inode_type(inode), index); + btrfs_inode_type(&inode->vfs_inode), index); if (ret == -EEXIST || ret == -EOVERFLOW) goto fail_dir_item; else if (ret) { @@ -6380,12 +6278,12 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, return ret; } - btrfs_i_size_write(parent_inode, parent_inode->i_size + + btrfs_i_size_write(parent_inode, parent_inode->vfs_inode.i_size + name_len * 2); - inode_inc_iversion(parent_inode); - parent_inode->i_mtime = parent_inode->i_ctime = - current_time(parent_inode); - ret = btrfs_update_inode(trans, root, parent_inode); + inode_inc_iversion(&parent_inode->vfs_inode); + parent_inode->vfs_inode.i_mtime = parent_inode->vfs_inode.i_ctime = + current_time(&parent_inode->vfs_inode); + ret = btrfs_update_inode(trans, root, &parent_inode->vfs_inode); if (ret) btrfs_abort_transaction(trans, ret); return ret; @@ -6394,9 +6292,9 @@ fail_dir_item: if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { u64 local_index; int err; - err = btrfs_del_root_ref(trans, root->fs_info->tree_root, - key.objectid, root->root_key.objectid, - parent_ino, &local_index, name, name_len); + err = btrfs_del_root_ref(trans, fs_info, key.objectid, + root->root_key.objectid, parent_ino, + &local_index, name, name_len); } else if (add_backref) { u64 local_index; @@ -6409,8 +6307,8 @@ fail_dir_item: } static int btrfs_add_nondir(struct btrfs_trans_handle *trans, - struct inode *dir, struct dentry *dentry, - struct inode *inode, int backref, u64 index) + struct btrfs_inode *dir, struct dentry *dentry, + struct btrfs_inode *inode, int backref, u64 index) { int err = btrfs_add_link(trans, dir, inode, dentry->d_name.name, dentry->d_name.len, @@ -6423,6 +6321,7 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans, static int btrfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; struct inode *inode = NULL; @@ -6445,8 +6344,8 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock; inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(dir), objectid, - mode, &index); + dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, + mode, &index); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto out_unlock; @@ -6465,7 +6364,8 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, if (err) goto out_unlock_inode; - err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode), + 0, index); if (err) { goto out_unlock_inode; } else { @@ -6475,9 +6375,9 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, } out_unlock: - btrfs_end_transaction(trans, root); - btrfs_balance_delayed_items(root); - btrfs_btree_balance_dirty(root); + btrfs_end_transaction(trans); + btrfs_balance_delayed_items(fs_info); + btrfs_btree_balance_dirty(fs_info); if (drop_inode) { inode_dec_link_count(inode); iput(inode); @@ -6494,6 +6394,7 @@ out_unlock_inode: static int btrfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; struct inode *inode = NULL; @@ -6516,8 +6417,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock; inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(dir), objectid, - mode, &index); + dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, + mode, &index); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto out_unlock; @@ -6541,7 +6442,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, if (err) goto out_unlock_inode; - err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode), + 0, index); if (err) goto out_unlock_inode; @@ -6550,13 +6452,13 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, d_instantiate(dentry, inode); out_unlock: - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (err && drop_inode_on_err) { inode_dec_link_count(inode); iput(inode); } - btrfs_balance_delayed_items(root); - btrfs_btree_balance_dirty(root); + btrfs_balance_delayed_items(fs_info); + btrfs_btree_balance_dirty(fs_info); return err; out_unlock_inode: @@ -6571,6 +6473,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, struct btrfs_trans_handle *trans = NULL; struct btrfs_root *root = BTRFS_I(dir)->root; struct inode *inode = d_inode(old_dentry); + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 index; int err; int drop_inode = 0; @@ -6582,7 +6485,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, if (inode->i_nlink >= BTRFS_LINK_MAX) return -EMLINK; - err = btrfs_set_inode_index(dir, &index); + err = btrfs_set_inode_index(BTRFS_I(dir), &index); if (err) goto fail; @@ -6606,7 +6509,8 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, ihold(inode); set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags); - err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index); + err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode), + 1, index); if (err) { drop_inode = 1; @@ -6620,28 +6524,29 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, * If new hard link count is 1, it's a file created * with open(2) O_TMPFILE flag. */ - err = btrfs_orphan_del(trans, inode); + err = btrfs_orphan_del(trans, BTRFS_I(inode)); if (err) goto fail; } d_instantiate(dentry, inode); - btrfs_log_new_name(trans, inode, NULL, parent); + btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent); } - btrfs_balance_delayed_items(root); + btrfs_balance_delayed_items(fs_info); fail: if (trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (drop_inode) { inode_dec_link_count(inode); iput(inode); } - btrfs_btree_balance_dirty(root); + btrfs_btree_balance_dirty(fs_info); return err; } static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct inode *inode = NULL; struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; @@ -6664,8 +6569,8 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto out_fail; inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(dir), objectid, - S_IFDIR | mode, &index); + dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, + S_IFDIR | mode, &index); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto out_fail; @@ -6680,13 +6585,14 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (err) goto out_fail_inode; - btrfs_i_size_write(inode, 0); + btrfs_i_size_write(BTRFS_I(inode), 0); err = btrfs_update_inode(trans, root, inode); if (err) goto out_fail_inode; - err = btrfs_add_link(trans, dir, inode, dentry->d_name.name, - dentry->d_name.len, 0, index); + err = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), + dentry->d_name.name, + dentry->d_name.len, 0, index); if (err) goto out_fail_inode; @@ -6699,13 +6605,13 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) drop_on_err = 0; out_fail: - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (drop_on_err) { inode_dec_link_count(inode); iput(inode); } - btrfs_balance_delayed_items(root); - btrfs_btree_balance_dirty(root); + btrfs_balance_delayed_items(fs_info); + btrfs_btree_balance_dirty(fs_info); return err; out_fail_inode: @@ -6803,6 +6709,20 @@ static noinline int uncompress_inline(struct btrfs_path *path, max_size = min_t(unsigned long, PAGE_SIZE, max_size); ret = btrfs_decompress(compress_type, tmp, page, extent_offset, inline_size, max_size); + + /* + * decompression code contains a memset to fill in any space between the end + * of the uncompressed data and the end of max_size in case the decompressed + * data ends up shorter than ram_bytes. That doesn't cover the hole between + * the end of an inline extent and the beginning of the next block, so we + * cover that region here. + */ + + if (max_size + pg_offset < PAGE_SIZE) { + char *map = kmap(page); + memset(map + pg_offset + max_size, 0, PAGE_SIZE - max_size - pg_offset); + kunmap(page); + } kfree(tmp); return ret; } @@ -6816,10 +6736,12 @@ static noinline int uncompress_inline(struct btrfs_path *path, * This also copies inline extents directly into the page. */ -struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page, - size_t pg_offset, u64 start, u64 len, - int create) +struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, + struct page *page, + size_t pg_offset, u64 start, u64 len, + int create) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); int ret; int err = 0; u64 extent_start = 0; @@ -6827,13 +6749,13 @@ struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page, u64 objectid = btrfs_ino(inode); u32 found_type; struct btrfs_path *path = NULL; - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; struct btrfs_file_extent_item *item; struct extent_buffer *leaf; struct btrfs_key found_key; struct extent_map *em = NULL; - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; + struct extent_map_tree *em_tree = &inode->extent_tree; + struct extent_io_tree *io_tree = &inode->io_tree; struct btrfs_trans_handle *trans = NULL; const bool new_inline = !page || create; @@ -6841,7 +6763,7 @@ again: read_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, start, len); if (em) - em->bdev = root->fs_info->fs_devices->latest_bdev; + em->bdev = fs_info->fs_devices->latest_bdev; read_unlock(&em_tree->lock); if (em) { @@ -6857,7 +6779,7 @@ again: err = -ENOMEM; goto out; } - em->bdev = root->fs_info->fs_devices->latest_bdev; + em->bdev = fs_info->fs_devices->latest_bdev; em->start = EXTENT_MAP_HOLE; em->orig_start = EXTENT_MAP_HOLE; em->len = (u64)-1; @@ -6916,7 +6838,8 @@ again: } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { size_t size; size = btrfs_file_extent_inline_len(leaf, path->slots[0], item); - extent_end = ALIGN(extent_start + size, root->sectorsize); + extent_end = ALIGN(extent_start + size, + fs_info->sectorsize); } next: if (start >= extent_end) { @@ -6945,7 +6868,8 @@ next: goto not_found_em; } - btrfs_extent_item_to_extent_map(inode, path, item, new_inline, em); + btrfs_extent_item_to_extent_map(inode, path, item, + new_inline, em); if (found_type == BTRFS_FILE_EXTENT_REG || found_type == BTRFS_FILE_EXTENT_PREALLOC) { @@ -6965,7 +6889,7 @@ next: copy_size = min_t(u64, PAGE_SIZE - pg_offset, size - extent_offset); em->start = extent_start + extent_offset; - em->len = ALIGN(copy_size, root->sectorsize); + em->len = ALIGN(copy_size, fs_info->sectorsize); em->orig_block_len = em->len; em->orig_start = em->start; ptr = btrfs_file_extent_inline_start(item) + extent_offset; @@ -7024,7 +6948,7 @@ not_found_em: insert: btrfs_release_path(path); if (em->start > start || extent_map_end(em) <= start) { - btrfs_err(root->fs_info, + btrfs_err(fs_info, "bad extent! em: [%llu %llu] passed [%llu %llu]", em->start, em->len, start, len); err = -EIO; @@ -7049,11 +6973,11 @@ insert: * extent causing the -EEXIST. */ if (existing->start == em->start && - extent_map_end(existing) == extent_map_end(em) && + extent_map_end(existing) >= extent_map_end(em) && em->block_start == existing->block_start) { /* - * these two extents are the same, it happens - * with inlines especially + * The existing extent map already encompasses the + * entire extent map we tried to add. */ free_extent_map(em); em = existing; @@ -7081,11 +7005,11 @@ insert: write_unlock(&em_tree->lock); out: - trace_btrfs_get_extent(root, em); + trace_btrfs_get_extent(root, inode, em); btrfs_free_path(path); if (trans) { - ret = btrfs_end_transaction(trans, root); + ret = btrfs_end_transaction(trans); if (!err) err = ret; } @@ -7097,9 +7021,10 @@ out: return em; } -struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page, - size_t pg_offset, u64 start, u64 len, - int create) +struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode, + struct page *page, + size_t pg_offset, u64 start, u64 len, + int create) { struct extent_map *em; struct extent_map *hole_em = NULL; @@ -7136,7 +7061,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *pag em = NULL; /* ok, we didn't find anything, lets look for delalloc */ - found = count_range_bits(&BTRFS_I(inode)->io_tree, &range_start, + found = count_range_bits(&inode->io_tree, &range_start, end, len, EXTENT_DELALLOC, 1); found_end = range_start + found; if (found_end < range_start) @@ -7237,11 +7162,12 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, struct extent_map *em = NULL; int ret; - down_read(&BTRFS_I(inode)->dio_sem); if (type != BTRFS_ORDERED_NOCOW) { - em = create_pinned_em(inode, start, len, orig_start, - block_start, block_len, orig_block_len, - ram_bytes, type); + em = create_io_em(inode, start, len, orig_start, + block_start, block_len, orig_block_len, + ram_bytes, + BTRFS_COMPRESS_NONE, /* compress_type */ + type); if (IS_ERR(em)) goto out; } @@ -7250,13 +7176,12 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, if (ret) { if (em) { free_extent_map(em); - btrfs_drop_extent_cache(inode, start, + btrfs_drop_extent_cache(BTRFS_I(inode), start, start + len - 1, 0); } em = ERR_PTR(ret); } out: - up_read(&BTRFS_I(inode)->dio_sem); return em; } @@ -7264,6 +7189,7 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, static struct extent_map *btrfs_new_extent_direct(struct inode *inode, u64 start, u64 len) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_map *em; struct btrfs_key ins; @@ -7271,17 +7197,18 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, int ret; alloc_hint = get_extent_allocation_hint(inode, start, len); - ret = btrfs_reserve_extent(root, len, len, root->sectorsize, 0, - alloc_hint, &ins, 1, 1); + ret = btrfs_reserve_extent(root, len, len, fs_info->sectorsize, + 0, alloc_hint, &ins, 1, 1); if (ret) return ERR_PTR(ret); em = btrfs_create_dio_extent(inode, start, ins.offset, start, ins.objectid, ins.offset, ins.offset, - ins.offset, 0); - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); + ins.offset, BTRFS_ORDERED_REGULAR); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); if (IS_ERR(em)) - btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 1); + btrfs_free_reserved_extent(fs_info, ins.objectid, + ins.offset, 1); return em; } @@ -7294,7 +7221,7 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, u64 *orig_start, u64 *orig_block_len, u64 *ram_bytes) { - struct btrfs_trans_handle *trans; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_path *path; int ret; struct extent_buffer *leaf; @@ -7314,8 +7241,8 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, if (!path) return -ENOMEM; - ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), - offset, 0); + ret = btrfs_lookup_file_extent(NULL, root, path, + btrfs_ino(BTRFS_I(inode)), offset, 0); if (ret < 0) goto out; @@ -7331,7 +7258,7 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, ret = 0; leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, slot); - if (key.objectid != btrfs_ino(inode) || + if (key.objectid != btrfs_ino(BTRFS_I(inode)) || key.type != BTRFS_EXTENT_DATA_KEY) { /* not our file or wrong item type, must cow */ goto out; @@ -7374,14 +7301,15 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); } - if (btrfs_extent_readonly(root, disk_bytenr)) + if (btrfs_extent_readonly(fs_info, disk_bytenr)) goto out; num_bytes = min(offset + *len, extent_end) - offset; if (!nocow && found_type == BTRFS_FILE_EXTENT_PREALLOC) { u64 range_end; - range_end = round_up(offset + num_bytes, root->sectorsize) - 1; + range_end = round_up(offset + num_bytes, + root->fs_info->sectorsize) - 1; ret = test_range_bit(io_tree, offset, range_end, EXTENT_DELALLOC, 0, NULL); if (ret) { @@ -7396,15 +7324,9 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, * look for other files referencing this extent, if we * find any we must cow */ - trans = btrfs_join_transaction(root); - if (IS_ERR(trans)) { - ret = 0; - goto out; - } - ret = btrfs_cross_ref_exist(trans, root, btrfs_ino(inode), + ret = btrfs_cross_ref_exist(root, btrfs_ino(BTRFS_I(inode)), key.offset - backref_offset, disk_bytenr); - btrfs_end_transaction(trans, root); if (ret) { ret = 0; goto out; @@ -7418,8 +7340,8 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, */ disk_bytenr += backref_offset; disk_bytenr += offset - key.offset; - if (csum_exist_in_range(root, disk_bytenr, num_bytes)) - goto out; + if (csum_exist_in_range(fs_info, disk_bytenr, num_bytes)) + goto out; /* * all of the above have passed, it is safe to overwrite this extent * without cow @@ -7515,7 +7437,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, * doing DIO to, so we need to make sure there's no ordered * extents in this range. */ - ordered = btrfs_lookup_ordered_range(inode, lockstart, + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), lockstart, lockend - lockstart + 1); /* @@ -7581,17 +7503,23 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, return ret; } -static struct extent_map *create_pinned_em(struct inode *inode, u64 start, - u64 len, u64 orig_start, - u64 block_start, u64 block_len, - u64 orig_block_len, u64 ram_bytes, - int type) +/* The callers of this must take lock_extent() */ +static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, + u64 orig_start, u64 block_start, + u64 block_len, u64 orig_block_len, + u64 ram_bytes, int compress_type, + int type) { struct extent_map_tree *em_tree; struct extent_map *em; struct btrfs_root *root = BTRFS_I(inode)->root; int ret; + ASSERT(type == BTRFS_ORDERED_PREALLOC || + type == BTRFS_ORDERED_COMPRESSED || + type == BTRFS_ORDERED_NOCOW || + type == BTRFS_ORDERED_REGULAR); + em_tree = &BTRFS_I(inode)->extent_tree; em = alloc_extent_map(); if (!em) @@ -7599,8 +7527,6 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start, em->start = start; em->orig_start = orig_start; - em->mod_start = start; - em->mod_len = len; em->len = len; em->block_len = block_len; em->block_start = block_start; @@ -7609,15 +7535,23 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start, em->ram_bytes = ram_bytes; em->generation = -1; set_bit(EXTENT_FLAG_PINNED, &em->flags); - if (type == BTRFS_ORDERED_PREALLOC) + if (type == BTRFS_ORDERED_PREALLOC) { set_bit(EXTENT_FLAG_FILLING, &em->flags); + } else if (type == BTRFS_ORDERED_COMPRESSED) { + set_bit(EXTENT_FLAG_COMPRESSED, &em->flags); + em->compress_type = compress_type; + } do { - btrfs_drop_extent_cache(inode, em->start, + btrfs_drop_extent_cache(BTRFS_I(inode), em->start, em->start + em->len - 1, 0); write_lock(&em_tree->lock); ret = add_extent_mapping(em_tree, em, 1); write_unlock(&em_tree->lock); + /* + * The caller has taken lock_extent(), who could race with us + * to add em? + */ } while (ret == -EEXIST); if (ret) { @@ -7625,6 +7559,7 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start, return ERR_PTR(ret); } + /* em got 2 refs now, callers needs to do free_extent_map once. */ return em; } @@ -7632,20 +7567,25 @@ static void adjust_dio_outstanding_extents(struct inode *inode, struct btrfs_dio_data *dio_data, const u64 len) { - unsigned num_extents; + unsigned num_extents = count_max_extents(len); - num_extents = (unsigned) div64_u64(len + BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); /* * If we have an outstanding_extents count still set then we're * within our reservation, otherwise we need to adjust our inode * counter appropriately. */ - if (dio_data->outstanding_extents) { + if (dio_data->outstanding_extents >= num_extents) { dio_data->outstanding_extents -= num_extents; } else { + /* + * If dio write length has been split due to no large enough + * contiguous space, we need to compensate our inode counter + * appropriately. + */ + u64 num_needed = num_extents - dio_data->outstanding_extents; + spin_lock(&BTRFS_I(inode)->lock); - BTRFS_I(inode)->outstanding_extents += num_extents; + BTRFS_I(inode)->outstanding_extents += num_needed; spin_unlock(&BTRFS_I(inode)->lock); } } @@ -7653,8 +7593,8 @@ static void adjust_dio_outstanding_extents(struct inode *inode, static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_map *em; - struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_state *cached_state = NULL; struct btrfs_dio_data *dio_data = NULL; u64 start = iblock << inode->i_blkbits; @@ -7666,7 +7606,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, if (create) unlock_bits |= EXTENT_DIRTY; else - len = min_t(u64, len, root->sectorsize); + len = min_t(u64, len, fs_info->sectorsize); lockstart = start; lockend = start + len - 1; @@ -7691,7 +7631,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, goto err; } - em = btrfs_get_extent(inode, NULL, 0, start, len, 0); + em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0); if (IS_ERR(em)) { ret = PTR_ERR(em); goto unlock_err; @@ -7755,14 +7695,14 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, if (can_nocow_extent(inode, start, &len, &orig_start, &orig_block_len, &ram_bytes) == 1 && - btrfs_inc_nocow_writers(root->fs_info, block_start)) { + btrfs_inc_nocow_writers(fs_info, block_start)) { struct extent_map *em2; em2 = btrfs_create_dio_extent(inode, start, len, orig_start, block_start, len, orig_block_len, ram_bytes, type); - btrfs_dec_nocow_writers(root->fs_info, block_start); + btrfs_dec_nocow_writers(fs_info, block_start); if (type == BTRFS_ORDERED_PREALLOC) { free_extent_map(em); em = em2; @@ -7808,7 +7748,7 @@ unlock: * Need to update the i_size under the extent lock so buffered * readers will get the updated i_size when we unlock. */ - if (start + len > i_size_read(inode)) + if (!dio_data->overwrite && start + len > i_size_read(inode)) i_size_write(inode, start + len); adjust_dio_outstanding_extents(inode, dio_data, len); @@ -7855,19 +7795,18 @@ err: static inline int submit_dio_repair_bio(struct inode *inode, struct bio *bio, int mirror_num) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int ret; BUG_ON(bio_op(bio) == REQ_OP_WRITE); bio_get(bio); - ret = btrfs_bio_wq_end_io(root->fs_info, bio, - BTRFS_WQ_ENDIO_DIO_REPAIR); + ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DIO_REPAIR); if (ret) goto err; - ret = btrfs_map_bio(root, bio, mirror_num, 0); + ret = btrfs_map_bio(fs_info, bio, mirror_num, 0); err: bio_put(bio); return ret; @@ -7917,7 +7856,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, struct io_failure_record *failrec; struct bio *bio; int isector; - int read_mode; + int read_mode = 0; int ret; BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE); @@ -7929,23 +7868,21 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, ret = btrfs_check_dio_repairable(inode, failed_bio, failrec, failed_mirror); if (!ret) { - free_io_failure(inode, failrec); + free_io_failure(BTRFS_I(inode), failrec); return -EIO; } if ((failed_bio->bi_vcnt > 1) || (failed_bio->bi_io_vec->bv_len - > BTRFS_I(inode)->root->sectorsize)) - read_mode = READ_SYNC | REQ_FAILFAST_DEV; - else - read_mode = READ_SYNC; + > btrfs_inode_sectorsize(inode))) + read_mode |= REQ_FAILFAST_DEV; isector = start - btrfs_io_bio(failed_bio)->logical; isector >>= inode->i_sb->s_blocksize_bits; bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page, pgoff, isector, repair_endio, repair_arg); if (!bio) { - free_io_failure(inode, failrec); + free_io_failure(BTRFS_I(inode), failrec); return -EIO; } bio_set_op_attrs(bio, REQ_OP_READ, read_mode); @@ -7956,7 +7893,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror); if (ret) { - free_io_failure(inode, failrec); + free_io_failure(BTRFS_I(inode), failrec); bio_put(bio); } @@ -7973,7 +7910,6 @@ struct btrfs_retry_complete { static void btrfs_retry_endio_nocsum(struct bio *bio) { struct btrfs_retry_complete *done = bio->bi_private; - struct inode *inode; struct bio_vec *bvec; int i; @@ -7981,12 +7917,12 @@ static void btrfs_retry_endio_nocsum(struct bio *bio) goto end; ASSERT(bio->bi_vcnt == 1); - inode = bio->bi_io_vec->bv_page->mapping->host; - ASSERT(bio->bi_io_vec->bv_len == BTRFS_I(inode)->root->sectorsize); + ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode)); done->uptodate = 1; bio_for_each_segment_all(bvec, bio, i) - clean_io_failure(done->inode, done->start, bvec->bv_page, 0); + clean_io_failure(BTRFS_I(done->inode), done->start, + bvec->bv_page, 0); end: complete(&done->done); bio_put(bio); @@ -8006,7 +7942,7 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, int ret; fs_info = BTRFS_I(inode)->root->fs_info; - sectorsize = BTRFS_I(inode)->root->sectorsize; + sectorsize = fs_info->sectorsize; start = io_bio->logical; done.inode = inode; @@ -8036,8 +7972,10 @@ next_block_or_try_again: start += sectorsize; - if (nr_sectors--) { + nr_sectors--; + if (nr_sectors) { pgoff += sectorsize; + ASSERT(pgoff < PAGE_SIZE); goto next_block_or_try_again; } } @@ -8049,9 +7987,7 @@ static void btrfs_retry_endio(struct bio *bio) { struct btrfs_retry_complete *done = bio->bi_private; struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); - struct inode *inode; struct bio_vec *bvec; - u64 start; int uptodate; int ret; int i; @@ -8061,18 +7997,15 @@ static void btrfs_retry_endio(struct bio *bio) uptodate = 1; - start = done->start; - ASSERT(bio->bi_vcnt == 1); - inode = bio->bi_io_vec->bv_page->mapping->host; - ASSERT(bio->bi_io_vec->bv_len == BTRFS_I(inode)->root->sectorsize); + ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode)); bio_for_each_segment_all(bvec, bio, i) { ret = __readpage_endio_check(done->inode, io_bio, i, bvec->bv_page, bvec->bv_offset, done->start, bvec->bv_len); if (!ret) - clean_io_failure(done->inode, done->start, + clean_io_failure(BTRFS_I(done->inode), done->start, bvec->bv_page, bvec->bv_offset); else uptodate = 0; @@ -8100,7 +8033,7 @@ static int __btrfs_subio_endio_read(struct inode *inode, int ret; fs_info = BTRFS_I(inode)->root->fs_info; - sectorsize = BTRFS_I(inode)->root->sectorsize; + sectorsize = fs_info->sectorsize; err = 0; start = io_bio->logical; @@ -8143,8 +8076,10 @@ next: ASSERT(nr_sectors); - if (--nr_sectors) { + nr_sectors--; + if (nr_sectors) { pgoff += sectorsize; + ASSERT(pgoff < PAGE_SIZE); goto next_block; } } @@ -8197,7 +8132,7 @@ static void btrfs_endio_direct_write_update_ordered(struct inode *inode, const u64 bytes, const int uptodate) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_ordered_extent *ordered = NULL; u64 ordered_offset = offset; u64 ordered_bytes = bytes; @@ -8213,8 +8148,7 @@ again: btrfs_init_work(&ordered->work, btrfs_endio_write_helper, finish_ordered_fn, NULL, NULL); - btrfs_queue_work(root->fs_info->endio_write_workers, - &ordered->work); + btrfs_queue_work(fs_info->endio_write_workers, &ordered->work); out_test: /* * our bio might span multiple ordered extents. If we haven't @@ -8249,8 +8183,7 @@ static int __btrfs_submit_bio_start_direct_io(struct inode *inode, unsigned long bio_flags, u64 offset) { int ret; - struct btrfs_root *root = BTRFS_I(inode)->root; - ret = btrfs_csum_one_bio(root, inode, bio, offset, 1); + ret = btrfs_csum_one_bio(inode, bio, offset, 1); BUG_ON(ret); /* -ENOMEM */ return 0; } @@ -8263,7 +8196,8 @@ static void btrfs_end_dio_bio(struct bio *bio) if (err) btrfs_warn(BTRFS_I(dip->inode)->root->fs_info, "direct IO failed ino %llu rw %d,%u sector %#Lx len %u err no %d", - btrfs_ino(dip->inode), bio_op(bio), bio->bi_opf, + btrfs_ino(BTRFS_I(dip->inode)), bio_op(bio), + bio->bi_opf, (unsigned long long)bio->bi_iter.bi_sector, bio->bi_iter.bi_size, err); @@ -8304,8 +8238,7 @@ static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev, return bio; } -static inline int btrfs_lookup_and_bind_dio_csum(struct btrfs_root *root, - struct inode *inode, +static inline int btrfs_lookup_and_bind_dio_csum(struct inode *inode, struct btrfs_dio_private *dip, struct bio *bio, u64 file_offset) @@ -8320,7 +8253,7 @@ static inline int btrfs_lookup_and_bind_dio_csum(struct btrfs_root *root, * contention. */ if (dip->logical_offset == file_offset) { - ret = btrfs_lookup_bio_sums_dio(root, inode, dip->orig_bio, + ret = btrfs_lookup_bio_sums_dio(inode, dip->orig_bio, file_offset); if (ret) return ret; @@ -8340,9 +8273,9 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, u64 file_offset, int skip_sum, int async_submit) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_dio_private *dip = bio->bi_private; bool write = bio_op(bio) == REQ_OP_WRITE; - struct btrfs_root *root = BTRFS_I(inode)->root; int ret; if (async_submit) @@ -8351,8 +8284,7 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, bio_get(bio); if (!write) { - ret = btrfs_bio_wq_end_io(root->fs_info, bio, - BTRFS_WQ_ENDIO_DATA); + ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA); if (ret) goto err; } @@ -8361,27 +8293,27 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, goto map; if (write && async_submit) { - ret = btrfs_wq_submit_bio(root->fs_info, - inode, bio, 0, 0, file_offset, - __btrfs_submit_bio_start_direct_io, - __btrfs_submit_bio_done); + ret = btrfs_wq_submit_bio(fs_info, inode, bio, 0, 0, + file_offset, + __btrfs_submit_bio_start_direct_io, + __btrfs_submit_bio_done); goto err; } else if (write) { /* * If we aren't doing async submit, calculate the csum of the * bio now. */ - ret = btrfs_csum_one_bio(root, inode, bio, file_offset, 1); + ret = btrfs_csum_one_bio(inode, bio, file_offset, 1); if (ret) goto err; } else { - ret = btrfs_lookup_and_bind_dio_csum(root, inode, dip, bio, + ret = btrfs_lookup_and_bind_dio_csum(inode, dip, bio, file_offset); if (ret) goto err; } map: - ret = btrfs_map_bio(root, bio, 0, async_submit); + ret = btrfs_map_bio(fs_info, bio, 0, async_submit); err: bio_put(bio); return ret; @@ -8391,23 +8323,24 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, int skip_sum) { struct inode *inode = dip->inode; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct bio *bio; struct bio *orig_bio = dip->orig_bio; - struct bio_vec *bvec = orig_bio->bi_io_vec; + struct bio_vec *bvec; u64 start_sector = orig_bio->bi_iter.bi_sector; u64 file_offset = dip->logical_offset; u64 submit_len = 0; u64 map_length; - u32 blocksize = root->sectorsize; + u32 blocksize = fs_info->sectorsize; int async_submit = 0; int nr_sectors; int ret; - int i; + int i, j; map_length = orig_bio->bi_iter.bi_size; - ret = btrfs_map_block(root->fs_info, bio_op(orig_bio), - start_sector << 9, &map_length, NULL, 0); + ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), start_sector << 9, + &map_length, NULL, 0); if (ret) return -EIO; @@ -8427,14 +8360,14 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, if (!bio) return -ENOMEM; - bio_set_op_attrs(bio, bio_op(orig_bio), bio_flags(orig_bio)); + bio->bi_opf = orig_bio->bi_opf; bio->bi_private = dip; bio->bi_end_io = btrfs_end_dio_bio; btrfs_io_bio(bio)->logical = file_offset; atomic_inc(&dip->pending_bios); - while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) { - nr_sectors = BTRFS_BYTES_TO_BLKS(root->fs_info, bvec->bv_len); + bio_for_each_segment_all(bvec, orig_bio, j) { + nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len); i = 0; next_block: if (unlikely(map_length < submit_len + blocksize || @@ -8465,14 +8398,13 @@ next_block: start_sector, GFP_NOFS); if (!bio) goto out_err; - bio_set_op_attrs(bio, bio_op(orig_bio), - bio_flags(orig_bio)); + bio->bi_opf = orig_bio->bi_opf; bio->bi_private = dip; bio->bi_end_io = btrfs_end_dio_bio; btrfs_io_bio(bio)->logical = file_offset; map_length = orig_bio->bi_iter.bi_size; - ret = btrfs_map_block(root->fs_info, bio_op(orig_bio), + ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), start_sector << 9, &map_length, NULL, 0); if (ret) { @@ -8487,7 +8419,6 @@ next_block: i++; goto next_block; } - bvec++; } } @@ -8619,12 +8550,13 @@ free_ordered: kfree(dip); } -static ssize_t check_direct_IO(struct btrfs_root *root, struct kiocb *iocb, - const struct iov_iter *iter, loff_t offset) +static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info, + struct kiocb *iocb, + const struct iov_iter *iter, loff_t offset) { int seg; int i; - unsigned blocksize_mask = root->sectorsize - 1; + unsigned int blocksize_mask = fs_info->sectorsize - 1; ssize_t retval = -EINVAL; if (offset & blocksize_mask) @@ -8656,7 +8588,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_dio_data dio_data = { 0 }; loff_t offset = iocb->ki_pos; size_t count = 0; @@ -8665,7 +8597,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) bool relock = false; ssize_t ret; - if (check_direct_IO(BTRFS_I(inode)->root, iocb, iter, offset)) + if (check_direct_IO(fs_info, iocb, iter, offset)) return 0; inode_dio_begin(inode); @@ -8690,25 +8622,26 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) * not unlock the i_mutex at this case. */ if (offset + count <= inode->i_size) { + dio_data.overwrite = 1; inode_unlock(inode); relock = true; } ret = btrfs_delalloc_reserve_space(inode, offset, count); if (ret) goto out; - dio_data.outstanding_extents = div64_u64(count + - BTRFS_MAX_EXTENT_SIZE - 1, - BTRFS_MAX_EXTENT_SIZE); + dio_data.outstanding_extents = count_max_extents(count); /* * We need to know how many extents we reserved so that we can * do the accounting properly if we go over the number we * originally calculated. Abuse current->journal_info for this. */ - dio_data.reserve = round_up(count, root->sectorsize); + dio_data.reserve = round_up(count, + fs_info->sectorsize); dio_data.unsubmitted_oe_range_start = (u64)offset; dio_data.unsubmitted_oe_range_end = (u64)offset; current->journal_info = &dio_data; + down_read(&BTRFS_I(inode)->dio_sem); } else if (test_bit(BTRFS_INODE_READDIO_NEED_LOCK, &BTRFS_I(inode)->runtime_flags)) { inode_dio_end(inode); @@ -8717,10 +8650,11 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) } ret = __blockdev_direct_IO(iocb, inode, - BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev, + fs_info->fs_devices->latest_bdev, iter, btrfs_get_blocks_direct, NULL, btrfs_submit_direct, flags); if (iov_iter_rw(iter) == WRITE) { + up_read(&BTRFS_I(inode)->dio_sem); current->journal_info = NULL; if (ret < 0 && ret != -EIOCBQUEUED) { if (dio_data.reserve) @@ -8839,7 +8773,7 @@ static int btrfs_releasepage(struct page *page, gfp_t gfp_flags) { if (PageWriteback(page) || PageDirty(page)) return 0; - return __btrfs_releasepage(page, gfp_flags & GFP_NOFS); + return __btrfs_releasepage(page, gfp_flags); } static void btrfs_invalidatepage(struct page *page, unsigned int offset, @@ -8874,7 +8808,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset, lock_extent_bits(tree, page_start, page_end, &cached_state); again: start = page_start; - ordered = btrfs_lookup_ordered_range(inode, start, + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start, page_end - start + 1); if (ordered) { end = min(page_end, ordered->file_offset + ordered->len - 1); @@ -8972,11 +8906,11 @@ again: * beyond EOF, then the page is guaranteed safe against truncation until we * unlock the page. */ -int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) +int btrfs_page_mkwrite(struct vm_fault *vmf) { struct page *page = vmf->page; - struct inode *inode = file_inode(vma->vm_file); - struct btrfs_root *root = BTRFS_I(inode)->root; + struct inode *inode = file_inode(vmf->vma->vm_file); + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; @@ -9008,7 +8942,7 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) ret = btrfs_delalloc_reserve_space(inode, page_start, reserved_space); if (!ret) { - ret = file_update_time(vma->vm_file); + ret = file_update_time(vmf->vma->vm_file); reserved = 1; } if (ret) { @@ -9040,7 +8974,8 @@ again: * we can't set the delalloc bits if there are pending ordered * extents. Drop our locks and wait for them to finish */ - ordered = btrfs_lookup_ordered_range(inode, page_start, page_end); + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start, + PAGE_SIZE); if (ordered) { unlock_extent_cached(io_tree, page_start, page_end, &cached_state, GFP_NOFS); @@ -9051,7 +8986,8 @@ again: } if (page->index == ((size - 1) >> PAGE_SHIFT)) { - reserved_space = round_up(size - page_start, root->sectorsize); + reserved_space = round_up(size - page_start, + fs_info->sectorsize); if (reserved_space < PAGE_SIZE) { end = page_start + reserved_space - 1; spin_lock(&BTRFS_I(inode)->lock); @@ -9063,11 +8999,11 @@ again: } /* - * XXX - page_mkwrite gets called every time the page is dirtied, even - * if it was already dirty, so for space accounting reasons we need to - * clear any delalloc bits for the range we are fixing to save. There - * is probably a better way to do this, but for now keep consistent with - * prepare_pages in the normal write path. + * page_mkwrite gets called when the page is firstly dirtied after it's + * faulted in, but write(2) could also dirty a page and set delalloc + * bits, thus in this case for space account reason, we still need to + * clear any delalloc bits within this page range since we have to + * reserve data&meta space before lock_page() (see above comments). */ clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end, EXTENT_DIRTY | EXTENT_DELALLOC | @@ -9100,7 +9036,7 @@ again: set_page_dirty(page); SetPageUptodate(page); - BTRFS_I(inode)->last_trans = root->fs_info->generation; + BTRFS_I(inode)->last_trans = fs_info->generation; BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid; BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->root->last_log_commit; @@ -9121,13 +9057,14 @@ out_noreserve: static int btrfs_truncate(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *rsv; int ret = 0; int err = 0; struct btrfs_trans_handle *trans; - u64 mask = root->sectorsize - 1; - u64 min_size = btrfs_calc_trunc_metadata_size(root, 1); + u64 mask = fs_info->sectorsize - 1; + u64 min_size = btrfs_calc_trunc_metadata_size(fs_info, 1); ret = btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1); @@ -9170,7 +9107,7 @@ static int btrfs_truncate(struct inode *inode) * 3) fs_info->trans_block_rsv - this will have 1 items worth left for * updating the inode. */ - rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); + rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); if (!rsv) return -ENOMEM; rsv->size = min_size; @@ -9187,7 +9124,7 @@ static int btrfs_truncate(struct inode *inode) } /* Migrate the slack space for the truncate to our reserve */ - ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv, + ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, 0); BUG_ON(ret); @@ -9210,15 +9147,15 @@ static int btrfs_truncate(struct inode *inode) break; } - trans->block_rsv = &root->fs_info->trans_block_rsv; + trans->block_rsv = &fs_info->trans_block_rsv; ret = btrfs_update_inode(trans, root, inode); if (ret) { err = ret; break; } - btrfs_end_transaction(trans, root); - btrfs_btree_balance_dirty(root); + btrfs_end_transaction(trans); + btrfs_btree_balance_dirty(fs_info); trans = btrfs_start_transaction(root, 2); if (IS_ERR(trans)) { @@ -9227,7 +9164,8 @@ static int btrfs_truncate(struct inode *inode) break; } - ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, + btrfs_block_rsv_release(fs_info, rsv, -1); + ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, 0); BUG_ON(ret); /* shouldn't happen */ trans->block_rsv = rsv; @@ -9235,22 +9173,22 @@ static int btrfs_truncate(struct inode *inode) if (ret == 0 && inode->i_nlink > 0) { trans->block_rsv = root->orphan_block_rsv; - ret = btrfs_orphan_del(trans, inode); + ret = btrfs_orphan_del(trans, BTRFS_I(inode)); if (ret) err = ret; } if (trans) { - trans->block_rsv = &root->fs_info->trans_block_rsv; + trans->block_rsv = &fs_info->trans_block_rsv; ret = btrfs_update_inode(trans, root, inode); if (ret && !err) err = ret; - ret = btrfs_end_transaction(trans, root); - btrfs_btree_balance_dirty(root); + ret = btrfs_end_transaction(trans); + btrfs_btree_balance_dirty(fs_info); } out: - btrfs_free_block_rsv(root, rsv); + btrfs_free_block_rsv(fs_info, rsv); if (ret && !err) err = ret; @@ -9280,7 +9218,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, inode->i_fop = &btrfs_dir_file_operations; set_nlink(inode, 1); - btrfs_i_size_write(inode, 0); + btrfs_i_size_write(BTRFS_I(inode), 0); unlock_new_inode(inode); err = btrfs_subvol_inherit_props(trans, new_root, parent_root); @@ -9353,7 +9291,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS void btrfs_test_destroy_inode(struct inode *inode) { - btrfs_drop_extent_cache(inode, 0, (u64)-1, 0); + btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0); kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); } #endif @@ -9366,6 +9304,7 @@ static void btrfs_i_callback(struct rcu_head *head) void btrfs_destroy_inode(struct inode *inode) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_ordered_extent *ordered; struct btrfs_root *root = BTRFS_I(inode)->root; @@ -9387,8 +9326,8 @@ void btrfs_destroy_inode(struct inode *inode) if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, &BTRFS_I(inode)->runtime_flags)) { - btrfs_info(root->fs_info, "inode %llu still on the orphan list", - btrfs_ino(inode)); + btrfs_info(fs_info, "inode %llu still on the orphan list", + btrfs_ino(BTRFS_I(inode))); atomic_dec(&root->orphan_inodes); } @@ -9397,7 +9336,7 @@ void btrfs_destroy_inode(struct inode *inode) if (!ordered) break; else { - btrfs_err(root->fs_info, + btrfs_err(fs_info, "found ordered extent %llu %llu on inode cleanup", ordered->file_offset, ordered->len); btrfs_remove_ordered_extent(inode, ordered); @@ -9407,7 +9346,7 @@ void btrfs_destroy_inode(struct inode *inode) } btrfs_qgroup_check_reserved_leak(inode); inode_tree_del(inode); - btrfs_drop_extent_cache(inode, 0, (u64)-1, 0); + btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0); free: call_rcu(&inode->i_rcu, btrfs_i_callback); } @@ -9486,11 +9425,11 @@ fail: return -ENOMEM; } -static int btrfs_getattr(struct vfsmount *mnt, - struct dentry *dentry, struct kstat *stat) +static int btrfs_getattr(const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int flags) { u64 delalloc_bytes; - struct inode *inode = d_inode(dentry); + struct inode *inode = d_inode(path->dentry); u32 blocksize = inode->i_sb->s_blocksize; generic_fillattr(inode, stat); @@ -9509,6 +9448,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, struct inode *new_dir, struct dentry *new_dentry) { + struct btrfs_fs_info *fs_info = btrfs_sb(old_dir->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(old_dir)->root; struct btrfs_root *dest = BTRFS_I(new_dir)->root; @@ -9516,8 +9456,8 @@ static int btrfs_rename_exchange(struct inode *old_dir, struct inode *old_inode = old_dentry->d_inode; struct timespec ctime = current_time(old_inode); struct dentry *parent; - u64 old_ino = btrfs_ino(old_inode); - u64 new_ino = btrfs_ino(new_inode); + u64 old_ino = btrfs_ino(BTRFS_I(old_inode)); + u64 new_ino = btrfs_ino(BTRFS_I(new_inode)); u64 old_idx = 0; u64 new_idx = 0; u64 root_objectid; @@ -9531,9 +9471,9 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* close the race window with snapshot create/destroy ioctl */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID) - down_read(&root->fs_info->subvol_sem); + down_read(&fs_info->subvol_sem); if (new_ino == BTRFS_FIRST_FREE_OBJECTID) - down_read(&dest->fs_info->subvol_sem); + down_read(&fs_info->subvol_sem); /* * We want to reserve the absolute worst case amount of items. So if @@ -9553,10 +9493,10 @@ static int btrfs_rename_exchange(struct inode *old_dir, * We need to find a free sequence number both in the source and * in the destination directory for the exchange. */ - ret = btrfs_set_inode_index(new_dir, &old_idx); + ret = btrfs_set_inode_index(BTRFS_I(new_dir), &old_idx); if (ret) goto out_fail; - ret = btrfs_set_inode_index(old_dir, &new_idx); + ret = btrfs_set_inode_index(BTRFS_I(old_dir), &new_idx); if (ret) goto out_fail; @@ -9566,7 +9506,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* Reference for the source. */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID) { /* force full log commit if subvolume involved. */ - btrfs_set_log_full_commit(root->fs_info, trans); + btrfs_set_log_full_commit(fs_info, trans); } else { btrfs_pin_log_trans(root); root_log_pinned = true; @@ -9574,7 +9514,8 @@ static int btrfs_rename_exchange(struct inode *old_dir, new_dentry->d_name.name, new_dentry->d_name.len, old_ino, - btrfs_ino(new_dir), old_idx); + btrfs_ino(BTRFS_I(new_dir)), + old_idx); if (ret) goto out_fail; } @@ -9582,7 +9523,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* And now for the dest. */ if (new_ino == BTRFS_FIRST_FREE_OBJECTID) { /* force full log commit if subvolume involved. */ - btrfs_set_log_full_commit(dest->fs_info, trans); + btrfs_set_log_full_commit(fs_info, trans); } else { btrfs_pin_log_trans(dest); dest_log_pinned = true; @@ -9590,7 +9531,8 @@ static int btrfs_rename_exchange(struct inode *old_dir, old_dentry->d_name.name, old_dentry->d_name.len, new_ino, - btrfs_ino(old_dir), new_idx); + btrfs_ino(BTRFS_I(old_dir)), + new_idx); if (ret) goto out_fail; } @@ -9606,8 +9548,10 @@ static int btrfs_rename_exchange(struct inode *old_dir, new_inode->i_ctime = ctime; if (old_dentry->d_parent != new_dentry->d_parent) { - btrfs_record_unlink_dir(trans, old_dir, old_inode, 1); - btrfs_record_unlink_dir(trans, new_dir, new_inode, 1); + btrfs_record_unlink_dir(trans, BTRFS_I(old_dir), + BTRFS_I(old_inode), 1); + btrfs_record_unlink_dir(trans, BTRFS_I(new_dir), + BTRFS_I(new_inode), 1); } /* src is a subvolume */ @@ -9618,8 +9562,8 @@ static int btrfs_rename_exchange(struct inode *old_dir, old_dentry->d_name.name, old_dentry->d_name.len); } else { /* src is an inode */ - ret = __btrfs_unlink_inode(trans, root, old_dir, - old_dentry->d_inode, + ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir), + BTRFS_I(old_dentry->d_inode), old_dentry->d_name.name, old_dentry->d_name.len); if (!ret) @@ -9638,8 +9582,8 @@ static int btrfs_rename_exchange(struct inode *old_dir, new_dentry->d_name.name, new_dentry->d_name.len); } else { /* dest is an inode */ - ret = __btrfs_unlink_inode(trans, dest, new_dir, - new_dentry->d_inode, + ret = __btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir), + BTRFS_I(new_dentry->d_inode), new_dentry->d_name.name, new_dentry->d_name.len); if (!ret) @@ -9650,7 +9594,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, goto out_fail; } - ret = btrfs_add_link(trans, new_dir, old_inode, + ret = btrfs_add_link(trans, BTRFS_I(new_dir), BTRFS_I(old_inode), new_dentry->d_name.name, new_dentry->d_name.len, 0, old_idx); if (ret) { @@ -9658,7 +9602,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, goto out_fail; } - ret = btrfs_add_link(trans, old_dir, new_inode, + ret = btrfs_add_link(trans, BTRFS_I(old_dir), BTRFS_I(new_inode), old_dentry->d_name.name, old_dentry->d_name.len, 0, new_idx); if (ret) { @@ -9673,13 +9617,15 @@ static int btrfs_rename_exchange(struct inode *old_dir, if (root_log_pinned) { parent = new_dentry->d_parent; - btrfs_log_new_name(trans, old_inode, old_dir, parent); + btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir), + parent); btrfs_end_log_trans(root); root_log_pinned = false; } if (dest_log_pinned) { parent = old_dentry->d_parent; - btrfs_log_new_name(trans, new_inode, new_dir, parent); + btrfs_log_new_name(trans, BTRFS_I(new_inode), BTRFS_I(new_dir), + parent); btrfs_end_log_trans(dest); dest_log_pinned = false; } @@ -9696,12 +9642,12 @@ out_fail: * allow the tasks to sync it. */ if (ret && (root_log_pinned || dest_log_pinned)) { - if (btrfs_inode_in_log(old_dir, root->fs_info->generation) || - btrfs_inode_in_log(new_dir, root->fs_info->generation) || - btrfs_inode_in_log(old_inode, root->fs_info->generation) || + if (btrfs_inode_in_log(BTRFS_I(old_dir), fs_info->generation) || + btrfs_inode_in_log(BTRFS_I(new_dir), fs_info->generation) || + btrfs_inode_in_log(BTRFS_I(old_inode), fs_info->generation) || (new_inode && - btrfs_inode_in_log(new_inode, root->fs_info->generation))) - btrfs_set_log_full_commit(root->fs_info, trans); + btrfs_inode_in_log(BTRFS_I(new_inode), fs_info->generation))) + btrfs_set_log_full_commit(fs_info, trans); if (root_log_pinned) { btrfs_end_log_trans(root); @@ -9712,12 +9658,12 @@ out_fail: dest_log_pinned = false; } } - ret = btrfs_end_transaction(trans, root); + ret = btrfs_end_transaction(trans); out_notrans: if (new_ino == BTRFS_FIRST_FREE_OBJECTID) - up_read(&dest->fs_info->subvol_sem); + up_read(&fs_info->subvol_sem); if (old_ino == BTRFS_FIRST_FREE_OBJECTID) - up_read(&root->fs_info->subvol_sem); + up_read(&fs_info->subvol_sem); return ret; } @@ -9739,7 +9685,7 @@ static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans, inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, dentry->d_name.len, - btrfs_ino(dir), + btrfs_ino(BTRFS_I(dir)), objectid, S_IFCHR | WHITEOUT_MODE, &index); @@ -9758,8 +9704,8 @@ static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans, if (ret) goto out; - ret = btrfs_add_nondir(trans, dir, dentry, - inode, 0, index); + ret = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, + BTRFS_I(inode), 0, index); if (ret) goto out; @@ -9777,6 +9723,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { + struct btrfs_fs_info *fs_info = btrfs_sb(old_dir->i_sb); struct btrfs_trans_handle *trans; unsigned int trans_num_items; struct btrfs_root *root = BTRFS_I(old_dir)->root; @@ -9786,10 +9733,10 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, u64 index = 0; u64 root_objectid; int ret; - u64 old_ino = btrfs_ino(old_inode); + u64 old_ino = btrfs_ino(BTRFS_I(old_inode)); bool log_pinned = false; - if (btrfs_ino(new_dir) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) + if (btrfs_ino(BTRFS_I(new_dir)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) return -EPERM; /* we only allow rename subvolume link between subvolumes */ @@ -9797,7 +9744,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, return -EXDEV; if (old_ino == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID || - (new_inode && btrfs_ino(new_inode) == BTRFS_FIRST_FREE_OBJECTID)) + (new_inode && btrfs_ino(BTRFS_I(new_inode)) == BTRFS_FIRST_FREE_OBJECTID)) return -ENOTEMPTY; if (S_ISDIR(old_inode->i_mode) && new_inode && @@ -9833,7 +9780,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, /* close the racy window with snapshot create/destroy ioctl */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID) - down_read(&root->fs_info->subvol_sem); + down_read(&fs_info->subvol_sem); /* * We want to reserve the absolute worst case amount of items. So if * both inodes are subvols and we need to unlink them then that would @@ -9857,14 +9804,14 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (dest != root) btrfs_record_root_in_trans(trans, dest); - ret = btrfs_set_inode_index(new_dir, &index); + ret = btrfs_set_inode_index(BTRFS_I(new_dir), &index); if (ret) goto out_fail; BTRFS_I(old_inode)->dir_index = 0ULL; if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) { /* force full log commit if subvolume involved. */ - btrfs_set_log_full_commit(root->fs_info, trans); + btrfs_set_log_full_commit(fs_info, trans); } else { btrfs_pin_log_trans(root); log_pinned = true; @@ -9872,7 +9819,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dentry->d_name.name, new_dentry->d_name.len, old_ino, - btrfs_ino(new_dir), index); + btrfs_ino(BTRFS_I(new_dir)), index); if (ret) goto out_fail; } @@ -9885,7 +9832,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, old_inode->i_ctime = current_time(old_dir); if (old_dentry->d_parent != new_dentry->d_parent) - btrfs_record_unlink_dir(trans, old_dir, old_inode, 1); + btrfs_record_unlink_dir(trans, BTRFS_I(old_dir), + BTRFS_I(old_inode), 1); if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) { root_objectid = BTRFS_I(old_inode)->root->root_key.objectid; @@ -9893,8 +9841,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, old_dentry->d_name.name, old_dentry->d_name.len); } else { - ret = __btrfs_unlink_inode(trans, root, old_dir, - d_inode(old_dentry), + ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir), + BTRFS_I(d_inode(old_dentry)), old_dentry->d_name.name, old_dentry->d_name.len); if (!ret) @@ -9908,7 +9856,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_inode) { inode_inc_iversion(new_inode); new_inode->i_ctime = current_time(new_inode); - if (unlikely(btrfs_ino(new_inode) == + if (unlikely(btrfs_ino(BTRFS_I(new_inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { root_objectid = BTRFS_I(new_inode)->location.objectid; ret = btrfs_unlink_subvol(trans, dest, new_dir, @@ -9917,20 +9865,21 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dentry->d_name.len); BUG_ON(new_inode->i_nlink == 0); } else { - ret = btrfs_unlink_inode(trans, dest, new_dir, - d_inode(new_dentry), + ret = btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir), + BTRFS_I(d_inode(new_dentry)), new_dentry->d_name.name, new_dentry->d_name.len); } if (!ret && new_inode->i_nlink == 0) - ret = btrfs_orphan_add(trans, d_inode(new_dentry)); + ret = btrfs_orphan_add(trans, + BTRFS_I(d_inode(new_dentry))); if (ret) { btrfs_abort_transaction(trans, ret); goto out_fail; } } - ret = btrfs_add_link(trans, new_dir, old_inode, + ret = btrfs_add_link(trans, BTRFS_I(new_dir), BTRFS_I(old_inode), new_dentry->d_name.name, new_dentry->d_name.len, 0, index); if (ret) { @@ -9944,7 +9893,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (log_pinned) { struct dentry *parent = new_dentry->d_parent; - btrfs_log_new_name(trans, old_inode, old_dir, parent); + btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir), + parent); btrfs_end_log_trans(root); log_pinned = false; } @@ -9971,20 +9921,20 @@ out_fail: * allow the tasks to sync it. */ if (ret && log_pinned) { - if (btrfs_inode_in_log(old_dir, root->fs_info->generation) || - btrfs_inode_in_log(new_dir, root->fs_info->generation) || - btrfs_inode_in_log(old_inode, root->fs_info->generation) || + if (btrfs_inode_in_log(BTRFS_I(old_dir), fs_info->generation) || + btrfs_inode_in_log(BTRFS_I(new_dir), fs_info->generation) || + btrfs_inode_in_log(BTRFS_I(old_inode), fs_info->generation) || (new_inode && - btrfs_inode_in_log(new_inode, root->fs_info->generation))) - btrfs_set_log_full_commit(root->fs_info, trans); + btrfs_inode_in_log(BTRFS_I(new_inode), fs_info->generation))) + btrfs_set_log_full_commit(fs_info, trans); btrfs_end_log_trans(root); log_pinned = false; } - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); out_notrans: if (old_ino == BTRFS_FIRST_FREE_OBJECTID) - up_read(&root->fs_info->subvol_sem); + up_read(&fs_info->subvol_sem); return ret; } @@ -10119,9 +10069,10 @@ out: int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) { + struct btrfs_fs_info *fs_info = root->fs_info; int ret; - if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state)) + if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) return -EROFS; ret = __start_delalloc_inodes(root, delay_iput, -1); @@ -10132,14 +10083,14 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) * we have to make sure the IO is actually started and that * ordered extents get created before we return */ - atomic_inc(&root->fs_info->async_submit_draining); - while (atomic_read(&root->fs_info->nr_async_submits) || - atomic_read(&root->fs_info->async_delalloc_pages)) { - wait_event(root->fs_info->async_submit_wait, - (atomic_read(&root->fs_info->nr_async_submits) == 0 && - atomic_read(&root->fs_info->async_delalloc_pages) == 0)); - } - atomic_dec(&root->fs_info->async_submit_draining); + atomic_inc(&fs_info->async_submit_draining); + while (atomic_read(&fs_info->nr_async_submits) || + atomic_read(&fs_info->async_delalloc_pages)) { + wait_event(fs_info->async_submit_wait, + (atomic_read(&fs_info->nr_async_submits) == 0 && + atomic_read(&fs_info->async_delalloc_pages) == 0)); + } + atomic_dec(&fs_info->async_submit_draining); return ret; } @@ -10202,6 +10153,7 @@ out: static int btrfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_path *path; @@ -10218,7 +10170,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, struct extent_buffer *leaf; name_len = strlen(symname); - if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(root)) + if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info)) return -ENAMETOOLONG; /* @@ -10237,8 +10189,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock; inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(dir), objectid, - S_IFLNK|S_IRWXUGO, &index); + dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), + objectid, S_IFLNK|S_IRWXUGO, &index); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto out_unlock; @@ -10264,7 +10216,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, err = -ENOMEM; goto out_unlock_inode; } - key.objectid = btrfs_ino(inode); + key.objectid = btrfs_ino(BTRFS_I(inode)); key.offset = 0; key.type = BTRFS_EXTENT_DATA_KEY; datasize = btrfs_file_extent_calc_inline_size(name_len); @@ -10294,7 +10246,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, inode_nohighmem(inode); inode->i_mapping->a_ops = &btrfs_symlink_aops; inode_set_bytes(inode, name_len); - btrfs_i_size_write(inode, name_len); + btrfs_i_size_write(BTRFS_I(inode), name_len); err = btrfs_update_inode(trans, root, inode); /* * Last step, add directory indexes for our symlink inode. This is the @@ -10302,7 +10254,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, * elsewhere above. */ if (!err) - err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, + BTRFS_I(inode), 0, index); if (err) { drop_inode = 1; goto out_unlock_inode; @@ -10312,12 +10265,12 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, d_instantiate(dentry, inode); out_unlock: - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (drop_inode) { inode_dec_link_count(inode); iput(inode); } - btrfs_btree_balance_dirty(root); + btrfs_btree_balance_dirty(fs_info); return err; out_unlock_inode: @@ -10331,6 +10284,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, loff_t actual_len, u64 *alloc_hint, struct btrfs_trans_handle *trans) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; struct extent_map *em; struct btrfs_root *root = BTRFS_I(inode)->root; @@ -10367,10 +10321,10 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, min_size, 0, *alloc_hint, &ins, 1, 0); if (ret) { if (own_trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); break; } - btrfs_dec_block_group_reservations(root->fs_info, ins.objectid); + btrfs_dec_block_group_reservations(fs_info, ins.objectid); last_alloc = ins.offset; ret = insert_reserved_file_extent(trans, inode, @@ -10379,15 +10333,15 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, ins.offset, 0, 0, 0, BTRFS_FILE_EXTENT_PREALLOC); if (ret) { - btrfs_free_reserved_extent(root, ins.objectid, + btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 0); btrfs_abort_transaction(trans, ret); if (own_trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); break; } - btrfs_drop_extent_cache(inode, cur_offset, + btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, cur_offset + ins.offset -1, 0); em = alloc_extent_map(); @@ -10404,7 +10358,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, 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; + em->bdev = fs_info->fs_devices->latest_bdev; set_bit(EXTENT_FLAG_PREALLOC, &em->flags); em->generation = trans->transid; @@ -10414,7 +10368,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, write_unlock(&em_tree->lock); if (ret != -EEXIST) break; - btrfs_drop_extent_cache(inode, cur_offset, + btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, cur_offset + ins.offset - 1, 0); } @@ -10443,12 +10397,12 @@ next: if (ret) { btrfs_abort_transaction(trans, ret); if (own_trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); break; } if (own_trans) - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); } if (cur_offset < end) btrfs_free_reserved_data_space(inode, cur_offset, @@ -10496,6 +10450,7 @@ static int btrfs_permission(struct inode *inode, int mask) static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) { + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; struct inode *inode = NULL; @@ -10515,7 +10470,7 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) goto out; inode = btrfs_new_inode(trans, root, dir, NULL, 0, - btrfs_ino(dir), objectid, mode, &index); + btrfs_ino(BTRFS_I(dir)), objectid, mode, &index); if (IS_ERR(inode)) { ret = PTR_ERR(inode); inode = NULL; @@ -10535,7 +10490,7 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) ret = btrfs_update_inode(trans, root, inode); if (ret) goto out_inode; - ret = btrfs_orphan_add(trans, inode); + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); if (ret) goto out_inode; @@ -10552,11 +10507,11 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) mark_inode_dirty(inode); out: - btrfs_end_transaction(trans, root); + btrfs_end_transaction(trans); if (ret) iput(inode); - btrfs_balance_delayed_items(root); - btrfs_btree_balance_dirty(root); + btrfs_balance_delayed_items(fs_info); + btrfs_btree_balance_dirty(fs_info); return ret; out_inode: @@ -10565,6 +10520,12 @@ out_inode: } +__attribute__((const)) +static int btrfs_readpage_io_failed_hook(struct page *page, int failed_mirror) +{ + return -EAGAIN; +} + static const struct inode_operations btrfs_dir_inode_operations = { .getattr = btrfs_getattr, .lookup = btrfs_lookup, @@ -10587,8 +10548,6 @@ static const struct inode_operations btrfs_dir_inode_operations = { static const struct inode_operations btrfs_dir_ro_inode_operations = { .lookup = btrfs_lookup, .permission = btrfs_permission, - .get_acl = btrfs_get_acl, - .set_acl = btrfs_set_acl, .update_time = btrfs_update_time, }; @@ -10605,10 +10564,14 @@ static const struct file_operations btrfs_dir_file_operations = { }; static const struct extent_io_ops btrfs_extent_io_ops = { - .fill_delalloc = run_delalloc_range, + /* mandatory callbacks */ .submit_bio_hook = btrfs_submit_bio_hook, - .merge_bio_hook = btrfs_merge_bio_hook, .readpage_end_io_hook = btrfs_readpage_end_io_hook, + .merge_bio_hook = btrfs_merge_bio_hook, + .readpage_io_failed_hook = btrfs_readpage_io_failed_hook, + + /* optional callbacks */ + .fill_delalloc = run_delalloc_range, .writepage_end_io_hook = btrfs_writepage_end_io_hook, .writepage_start_hook = btrfs_writepage_start_hook, .set_bit_hook = btrfs_set_bit_hook, @@ -10668,7 +10631,6 @@ static const struct inode_operations btrfs_special_inode_operations = { .update_time = btrfs_update_time, }; static const struct inode_operations btrfs_symlink_inode_operations = { - .readlink = generic_readlink, .get_link = page_get_link, .getattr = btrfs_getattr, .setattr = btrfs_setattr, |