diff options
Diffstat (limited to 'fs/btrfs/inode.c')
| -rw-r--r-- | fs/btrfs/inode.c | 8359 |
1 files changed, 4612 insertions, 3747 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d267eb5caa7b..0e516aefbf51 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3,9 +3,10 @@ * Copyright (C) 2007 Oracle. All rights reserved. */ +#include <crypto/hash.h> #include <linux/kernel.h> #include <linux/bio.h> -#include <linux/buffer_head.h> +#include <linux/blk-cgroup.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/pagemap.h> @@ -28,8 +29,11 @@ #include <linux/magic.h> #include <linux/iversion.h> #include <linux/swap.h> +#include <linux/migrate.h> #include <linux/sched/mm.h> +#include <linux/iomap.h> #include <asm/unaligned.h> +#include <linux/fsverity.h> #include "misc.h" #include "ctree.h" #include "disk-io.h" @@ -43,22 +47,56 @@ #include "compression.h" #include "locking.h" #include "free-space-cache.h" -#include "inode-map.h" #include "props.h" #include "qgroup.h" #include "delalloc-space.h" #include "block-group.h" +#include "space-info.h" +#include "zoned.h" +#include "subpage.h" +#include "inode-item.h" struct btrfs_iget_args { - struct btrfs_key *location; + u64 ino; struct btrfs_root *root; }; struct btrfs_dio_data { - u64 reserve; - u64 unsubmitted_oe_range_start; - u64 unsubmitted_oe_range_end; - int overwrite; + ssize_t submitted; + struct extent_changeset *data_reserved; + bool data_space_reserved; + bool nocow_done; +}; + +struct btrfs_dio_private { + struct inode *inode; + + /* + * Since DIO can use anonymous page, we cannot use page_offset() to + * grab the file offset, thus need a dedicated member for file offset. + */ + u64 file_offset; + /* Used for bio::bi_size */ + u32 bytes; + + /* + * References to this structure. There is one reference per in-flight + * bio plus one while we're still setting up. + */ + refcount_t refs; + + /* Array of checksums */ + u8 *csums; + + /* This must be last */ + struct bio bio; +}; + +static struct bio_set btrfs_dio_bioset; + +struct btrfs_rename_ctx { + /* Output field. Stores the index number of the old directory entry. */ + u64 index; }; static const struct inode_operations btrfs_dir_inode_operations; @@ -67,7 +105,6 @@ static const struct inode_operations btrfs_special_inode_operations; static const struct inode_operations btrfs_file_inode_operations; static const struct address_space_operations btrfs_aops; static const struct file_operations btrfs_dir_file_operations; -static const struct extent_io_ops btrfs_extent_io_ops; static struct kmem_cache *btrfs_inode_cachep; struct kmem_cache *btrfs_trans_handle_cachep; @@ -77,20 +114,66 @@ struct kmem_cache *btrfs_free_space_bitmap_cachep; static int btrfs_setsize(struct inode *inode, struct iattr *attr); static int btrfs_truncate(struct inode *inode, bool skip_writeback); -static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent); -static noinline int cow_file_range(struct inode *inode, +static noinline int cow_file_range(struct btrfs_inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, - unsigned long *nr_written, int unlock); -static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, - u64 orig_start, u64 block_start, + unsigned long *nr_written, int unlock, + u64 *done_offset); +static struct extent_map *create_io_em(struct btrfs_inode *inode, u64 start, + u64 len, u64 orig_start, u64 block_start, u64 block_len, u64 orig_block_len, u64 ram_bytes, int compress_type, int type); -static void __endio_write_update_ordered(struct inode *inode, - const u64 offset, const u64 bytes, - const bool uptodate); +/* + * btrfs_inode_lock - lock inode i_rwsem based on arguments passed + * + * ilock_flags can have the following bit set: + * + * BTRFS_ILOCK_SHARED - acquire a shared lock on the inode + * BTRFS_ILOCK_TRY - try to acquire the lock, if fails on first attempt + * return -EAGAIN + * BTRFS_ILOCK_MMAP - acquire a write lock on the i_mmap_lock + */ +int btrfs_inode_lock(struct inode *inode, unsigned int ilock_flags) +{ + if (ilock_flags & BTRFS_ILOCK_SHARED) { + if (ilock_flags & BTRFS_ILOCK_TRY) { + if (!inode_trylock_shared(inode)) + return -EAGAIN; + else + return 0; + } + inode_lock_shared(inode); + } else { + if (ilock_flags & BTRFS_ILOCK_TRY) { + if (!inode_trylock(inode)) + return -EAGAIN; + else + return 0; + } + inode_lock(inode); + } + if (ilock_flags & BTRFS_ILOCK_MMAP) + down_write(&BTRFS_I(inode)->i_mmap_lock); + return 0; +} + +/* + * btrfs_inode_unlock - unock inode i_rwsem + * + * ilock_flags should contain the same bits set as passed to btrfs_inode_lock() + * to decide whether the lock acquired is shared or exclusive. + */ +void btrfs_inode_unlock(struct inode *inode, unsigned int ilock_flags) +{ + if (ilock_flags & BTRFS_ILOCK_MMAP) + up_write(&BTRFS_I(inode)->i_mmap_lock); + if (ilock_flags & BTRFS_ILOCK_SHARED) + inode_unlock_shared(inode); + else + inode_unlock(inode); +} /* * Cleanup all submitted ordered extents in specified range to handle errors @@ -102,58 +185,92 @@ static void __endio_write_update_ordered(struct inode *inode, * to be released, which we want to happen only when finishing the ordered * extent (btrfs_finish_ordered_io()). */ -static inline void btrfs_cleanup_ordered_extents(struct inode *inode, +static inline void btrfs_cleanup_ordered_extents(struct btrfs_inode *inode, struct page *locked_page, u64 offset, u64 bytes) { unsigned long index = offset >> PAGE_SHIFT; unsigned long end_index = (offset + bytes - 1) >> PAGE_SHIFT; - u64 page_start = page_offset(locked_page); - u64 page_end = page_start + PAGE_SIZE - 1; - + u64 page_start, page_end; struct page *page; + if (locked_page) { + page_start = page_offset(locked_page); + page_end = page_start + PAGE_SIZE - 1; + } + while (index <= end_index) { - page = find_get_page(inode->i_mapping, index); + /* + * For locked page, we will call end_extent_writepage() on it + * in run_delalloc_range() for the error handling. That + * end_extent_writepage() function will call + * btrfs_mark_ordered_io_finished() to clear page Ordered and + * run the ordered extent accounting. + * + * Here we can't just clear the Ordered bit, or + * btrfs_mark_ordered_io_finished() would skip the accounting + * for the page range, and the ordered extent will never finish. + */ + if (locked_page && index == (page_start >> PAGE_SHIFT)) { + index++; + continue; + } + page = find_get_page(inode->vfs_inode.i_mapping, index); index++; if (!page) continue; - ClearPagePrivate2(page); + + /* + * Here we just clear all Ordered bits for every page in the + * range, then btrfs_mark_ordered_io_finished() will handle + * the ordered extent accounting for the range. + */ + btrfs_page_clamp_clear_ordered(inode->root->fs_info, page, + offset, bytes); put_page(page); } - /* - * In case this page belongs to the delalloc range being instantiated - * then skip it, since the first page of a range is going to be - * properly cleaned up by the caller of run_delalloc_range - */ - if (page_start >= offset && page_end <= (offset + bytes - 1)) { - offset += PAGE_SIZE; - bytes -= PAGE_SIZE; + if (locked_page) { + /* The locked page covers the full range, nothing needs to be done */ + if (bytes + offset <= page_start + PAGE_SIZE) + return; + /* + * In case this page belongs to the delalloc range being + * instantiated then skip it, since the first page of a range is + * going to be properly cleaned up by the caller of + * run_delalloc_range + */ + if (page_start >= offset && page_end <= (offset + bytes - 1)) { + bytes = offset + bytes - page_offset(locked_page) - PAGE_SIZE; + offset = page_offset(locked_page) + PAGE_SIZE; + } } - return __endio_write_update_ordered(inode, offset, bytes, false); + return btrfs_mark_ordered_io_finished(inode, NULL, offset, bytes, false); } static int btrfs_dirty_inode(struct inode *inode); -#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS -void btrfs_test_inode_set_ops(struct inode *inode) -{ - BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; -} -#endif - static int btrfs_init_inode_security(struct btrfs_trans_handle *trans, - struct inode *inode, struct inode *dir, - const struct qstr *qstr) + struct btrfs_new_inode_args *args) { int err; - err = btrfs_init_acl(trans, inode, dir); - if (!err) - err = btrfs_xattr_security_init(trans, inode, dir, qstr); - return err; + if (args->default_acl) { + err = __btrfs_set_acl(trans, args->inode, args->default_acl, + ACL_TYPE_DEFAULT); + if (err) + return err; + } + if (args->acl) { + err = __btrfs_set_acl(trans, args->inode, args->acl, ACL_TYPE_ACCESS); + if (err) + return err; + } + if (!args->default_acl && !args->acl) + cache_no_acl(args->inode); + return btrfs_xattr_security_init(trans, args->inode, args->dir, + &args->dentry->d_name); } /* @@ -162,12 +279,14 @@ static int btrfs_init_inode_security(struct btrfs_trans_handle *trans, * no overlapping inline items exist in the btree */ static int insert_inline_extent(struct btrfs_trans_handle *trans, - struct btrfs_path *path, int extent_inserted, - struct btrfs_root *root, struct inode *inode, - u64 start, size_t size, size_t compressed_size, + struct btrfs_path *path, + struct btrfs_inode *inode, bool extent_inserted, + size_t size, size_t compressed_size, int compress_type, - struct page **compressed_pages) + struct page **compressed_pages, + bool update_i_size) { + struct btrfs_root *root = inode->root; struct extent_buffer *leaf; struct page *page = NULL; char *kaddr; @@ -175,7 +294,7 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, struct btrfs_file_extent_item *ei; int ret; size_t cur_size = size; - unsigned long offset; + u64 i_size; ASSERT((compressed_size > 0 && compressed_pages) || (compressed_size == 0 && !compressed_pages)); @@ -183,18 +302,15 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, if (compressed_size && compressed_pages) cur_size = compressed_size; - inode_add_bytes(inode, size); - if (!extent_inserted) { struct btrfs_key key; size_t datasize; - key.objectid = btrfs_ino(BTRFS_I(inode)); - key.offset = start; + key.objectid = btrfs_ino(inode); + key.offset = 0; key.type = BTRFS_EXTENT_DATA_KEY; datasize = btrfs_file_extent_calc_inline_size(cur_size); - path->leave_spinning = 1; ret = btrfs_insert_empty_item(trans, root, path, &key, datasize); if (ret) @@ -218,9 +334,9 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, cur_size = min_t(unsigned long, compressed_size, PAGE_SIZE); - kaddr = kmap_atomic(cpage); + kaddr = kmap_local_page(cpage); write_extent_buffer(leaf, kaddr, ptr, cur_size); - kunmap_atomic(kaddr); + kunmap_local(kaddr); i++; ptr += cur_size; @@ -229,29 +345,38 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, btrfs_set_file_extent_compression(leaf, ei, compress_type); } else { - page = find_get_page(inode->i_mapping, - start >> PAGE_SHIFT); + page = find_get_page(inode->vfs_inode.i_mapping, 0); btrfs_set_file_extent_compression(leaf, ei, 0); - kaddr = kmap_atomic(page); - offset = offset_in_page(start); - write_extent_buffer(leaf, kaddr + offset, ptr, size); - kunmap_atomic(kaddr); + kaddr = kmap_local_page(page); + write_extent_buffer(leaf, kaddr, ptr, size); + kunmap_local(kaddr); put_page(page); } btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); /* - * we're an inline extent, so nobody can - * extend the file past i_size without locking - * a page we already have locked. + * We align size to sectorsize for inline extents just for simplicity + * sake. + */ + ret = btrfs_inode_set_file_extent_range(inode, 0, + ALIGN(size, root->fs_info->sectorsize)); + if (ret) + goto fail; + + /* + * We're an inline extent, so nobody can extend the file past i_size + * without locking a page we already have locked. * - * We must do any isize and inode updates - * before we unlock the pages. Otherwise we - * could end up racing with unlink. + * We must do any i_size and inode updates before we unlock the pages. + * Otherwise we could end up racing with unlink. */ - BTRFS_I(inode)->disk_i_size = inode->i_size; - ret = btrfs_update_inode(trans, root, inode); + i_size = i_size_read(&inode->vfs_inode); + if (update_i_size && size > i_size) { + i_size_write(&inode->vfs_inode, size); + i_size = size; + } + inode->disk_i_size = i_size; fail: return ret; @@ -263,36 +388,31 @@ fail: * does the checks required to make sure the data is small enough * to fit as an inline extent. */ -static noinline int cow_file_range_inline(struct inode *inode, u64 start, - u64 end, size_t compressed_size, +static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, + size_t compressed_size, int compress_type, - struct page **compressed_pages) + struct page **compressed_pages, + bool update_i_size) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_drop_extents_args drop_args = { 0 }; + struct btrfs_root *root = inode->root; 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, fs_info->sectorsize); - u64 data_len = inline_len; + u64 data_len = (compressed_size ?: size); int ret; struct btrfs_path *path; - int extent_inserted = 0; - u32 extent_item_size; - - if (compressed_size) - data_len = compressed_size; - if (start > 0 || - actual_end > fs_info->sectorsize || + /* + * We can create an inline extent if it ends at or beyond the current + * i_size, is no larger than a sector (decompressed), and the (possibly + * compressed) data fits in a leaf and the configured maximum inline + * size. + */ + if (size < i_size_read(&inode->vfs_inode) || + size > fs_info->sectorsize || data_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info) || - (!compressed_size && - (actual_end & (fs_info->sectorsize - 1)) == 0) || - end + 1 < isize || - data_len > fs_info->max_inline) { + data_len > fs_info->max_inline) return 1; - } path = btrfs_alloc_path(); if (!path) @@ -303,29 +423,23 @@ static noinline int cow_file_range_inline(struct inode *inode, u64 start, btrfs_free_path(path); return PTR_ERR(trans); } - trans->block_rsv = &BTRFS_I(inode)->block_rsv; - - if (compressed_size && compressed_pages) - extent_item_size = btrfs_file_extent_calc_inline_size( - compressed_size); - else - extent_item_size = btrfs_file_extent_calc_inline_size( - inline_len); + trans->block_rsv = &inode->block_rsv; - ret = __btrfs_drop_extents(trans, root, inode, path, - start, aligned_end, NULL, - 1, 1, extent_item_size, &extent_inserted); + drop_args.path = path; + drop_args.start = 0; + drop_args.end = fs_info->sectorsize; + drop_args.drop_cache = true; + drop_args.replace_extent = true; + drop_args.extent_item_size = btrfs_file_extent_calc_inline_size(data_len); + ret = btrfs_drop_extents(trans, root, inode, &drop_args); if (ret) { btrfs_abort_transaction(trans, ret); goto out; } - if (isize > actual_end) - inline_len = min_t(u64, isize, actual_end); - ret = insert_inline_extent(trans, path, extent_inserted, - root, inode, start, - inline_len, compressed_size, - compress_type, compressed_pages); + ret = insert_inline_extent(trans, path, inode, drop_args.extent_inserted, + size, compressed_size, compress_type, + compressed_pages, update_i_size); if (ret && ret != -ENOSPC) { btrfs_abort_transaction(trans, ret); goto out; @@ -334,8 +448,17 @@ static noinline int cow_file_range_inline(struct inode *inode, u64 start, goto out; } - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); - btrfs_drop_extent_cache(BTRFS_I(inode), start, aligned_end - 1, 0); + btrfs_update_inode_bytes(inode, size, drop_args.bytes_found); + ret = btrfs_update_inode(trans, root, inode); + if (ret && ret != -ENOSPC) { + btrfs_abort_transaction(trans, ret); + goto out; + } else if (ret == -ENOSPC) { + ret = 1; + goto out; + } + + btrfs_set_inode_full_sync(inode); out: /* * Don't forget to free the reserved space, as for inlined extent @@ -364,15 +487,14 @@ struct async_chunk { struct page *locked_page; u64 start; u64 end; - unsigned int write_flags; + blk_opf_t write_flags; struct list_head extents; struct cgroup_subsys_state *blkcg_css; struct btrfs_work work; - atomic_t *pending; + struct async_cow *async_cow; }; struct async_cow { - /* Number of chunks in flight; must be first in the structure */ atomic_t num_chunks; struct async_chunk chunks[]; }; @@ -399,53 +521,75 @@ static noinline int add_async_extent(struct async_chunk *cow, } /* - * Check if the inode has flags compatible with compression - */ -static inline bool inode_can_compress(struct inode *inode) -{ - if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW || - BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) - return false; - return true; -} - -/* * Check if the inode needs to be submitted to compression, based on mount * options, defragmentation, properties or heuristics. */ -static inline int inode_need_compress(struct inode *inode, u64 start, u64 end) +static inline int inode_need_compress(struct btrfs_inode *inode, u64 start, + u64 end) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct btrfs_fs_info *fs_info = inode->root->fs_info; - if (!inode_can_compress(inode)) { + if (!btrfs_inode_can_compress(inode)) { WARN(IS_ENABLED(CONFIG_BTRFS_DEBUG), KERN_ERR "BTRFS: unexpected compression for ino %llu\n", - btrfs_ino(BTRFS_I(inode))); + btrfs_ino(inode)); return 0; } + /* + * Special check for subpage. + * + * We lock the full page then run each delalloc range in the page, thus + * for the following case, we will hit some subpage specific corner case: + * + * 0 32K 64K + * | |///////| |///////| + * \- A \- B + * + * In above case, both range A and range B will try to unlock the full + * page [0, 64K), causing the one finished later will have page + * unlocked already, triggering various page lock requirement BUG_ON()s. + * + * So here we add an artificial limit that subpage compression can only + * if the range is fully page aligned. + * + * In theory we only need to ensure the first page is fully covered, but + * the tailing partial page will be locked until the full compression + * finishes, delaying the write of other range. + * + * TODO: Make btrfs_run_delalloc_range() to lock all delalloc range + * first to prevent any submitted async extent to unlock the full page. + * By this, we can ensure for subpage case that only the last async_cow + * will unlock the full page. + */ + if (fs_info->sectorsize < PAGE_SIZE) { + if (!PAGE_ALIGNED(start) || + !PAGE_ALIGNED(end + 1)) + return 0; + } + /* force compress */ if (btrfs_test_opt(fs_info, FORCE_COMPRESS)) return 1; /* defrag ioctl */ - if (BTRFS_I(inode)->defrag_compress) + if (inode->defrag_compress) return 1; /* bad compression ratios */ - if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS) + if (inode->flags & BTRFS_INODE_NOCOMPRESS) return 0; if (btrfs_test_opt(fs_info, COMPRESS) || - BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS || - BTRFS_I(inode)->prop_compress) - return btrfs_compress_heuristic(inode, start, end); + inode->flags & BTRFS_INODE_COMPRESS || + inode->prop_compress) + return btrfs_compress_heuristic(&inode->vfs_inode, start, end); return 0; } static inline void inode_should_defrag(struct btrfs_inode *inode, - u64 start, u64 end, u64 num_bytes, u64 small_write) + u64 start, u64 end, u64 num_bytes, u32 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); + btrfs_add_inode_defrag(NULL, inode, small_write); } /* @@ -504,7 +648,6 @@ static noinline int compress_file_range(struct async_chunk *async_chunk) again: will_compress = 0; nr_pages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1; - BUILD_BUG_ON((BTRFS_MAX_COMPRESSED % PAGE_SIZE) != 0); nr_pages = min_t(unsigned long, nr_pages, BTRFS_MAX_COMPRESSED / PAGE_SIZE); @@ -524,13 +667,24 @@ again: total_compressed = actual_end - start; /* - * skip compression for a small file range(<=blocksize) that + * Skip compression for a small file range(<=blocksize) that * isn't an inline extent, since it doesn't save disk space at all. */ if (total_compressed <= blocksize && (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size)) goto cleanup_and_bail_uncompressed; + /* + * For subpage case, we require full page alignment for the sector + * aligned range. + * Thus we must also check against @actual_end, not just @end. + */ + if (blocksize < PAGE_SIZE) { + if (!PAGE_ALIGNED(start) || + !PAGE_ALIGNED(round_up(actual_end, blocksize))) + goto cleanup_and_bail_uncompressed; + } + total_compressed = min_t(unsigned long, total_compressed, BTRFS_MAX_UNCOMPRESSED); total_in = 0; @@ -541,7 +695,7 @@ again: * inode has not been flagged as nocompress. This flag can * change at any time if we discover bad compression ratios. */ - if (inode_need_compress(inode, start, end)) { + if (inode_need_compress(BTRFS_I(inode), start, end)) { WARN_ON(pages); pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS); if (!pages) { @@ -584,34 +738,35 @@ again: if (!ret) { unsigned long offset = offset_in_page(total_compressed); struct page *page = pages[nr_pages - 1]; - char *kaddr; /* zero the tail end of the last page, we might be * sending it down to disk */ - if (offset) { - kaddr = kmap_atomic(page); - memset(kaddr + offset, 0, - PAGE_SIZE - offset); - kunmap_atomic(kaddr); - } + if (offset) + memzero_page(page, offset, PAGE_SIZE - offset); will_compress = 1; } } cont: - if (start == 0) { + /* + * Check cow_file_range() for why we don't even try to create inline + * extent for subpage case. + */ + if (start == 0 && fs_info->sectorsize == PAGE_SIZE) { /* lets try to make an inline extent */ if (ret || total_in < actual_end) { /* we didn't compress the entire range, try * to make an uncompressed inline extent. */ - ret = cow_file_range_inline(inode, start, end, 0, - BTRFS_COMPRESS_NONE, NULL); + ret = cow_file_range_inline(BTRFS_I(inode), actual_end, + 0, BTRFS_COMPRESS_NONE, + NULL, false); } else { /* try making a compressed inline extent */ - ret = cow_file_range_inline(inode, start, end, + ret = cow_file_range_inline(BTRFS_I(inode), actual_end, total_compressed, - compress_type, pages); + compress_type, pages, + false); } if (ret <= 0) { unsigned long clear_flags = EXTENT_DELALLOC | @@ -631,20 +786,26 @@ cont: * our outstanding extent for clearing delalloc for this * range. */ - extent_clear_unlock_delalloc(inode, start, end, NULL, + extent_clear_unlock_delalloc(BTRFS_I(inode), start, end, + NULL, clear_flags, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | page_error_op | PAGE_END_WRITEBACK); - for (i = 0; i < nr_pages; i++) { - WARN_ON(pages[i]->mapping); - put_page(pages[i]); + /* + * Ensure we only free the compressed pages if we have + * them allocated, as we can still reach here with + * inode_need_compress() == false. + */ + if (pages) { + for (i = 0; i < nr_pages; i++) { + WARN_ON(pages[i]->mapping); + put_page(pages[i]); + } + kfree(pages); } - kfree(pages); - return 0; } } @@ -662,7 +823,7 @@ cont: * win, compare the page count read with the blocks on disk, * compression must free at least one sector size */ - total_in = ALIGN(total_in, PAGE_SIZE); + total_in = round_up(total_in, fs_info->sectorsize); if (total_compressed + blocksize <= total_in) { compressed_extents++; @@ -743,185 +904,207 @@ static void free_async_extent_pages(struct async_extent *async_extent) async_extent->pages = NULL; } -/* - * phase two of compressed writeback. This is the ordered portion - * of the code, which only gets called in the order the work was - * queued. We walk all the async extents created by compress_file_range - * and send them down to the disk. - */ -static noinline void submit_compressed_extents(struct async_chunk *async_chunk) +static int submit_uncompressed_range(struct btrfs_inode *inode, + struct async_extent *async_extent, + struct page *locked_page) { - struct inode *inode = async_chunk->inode; - 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_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - int ret = 0; + u64 start = async_extent->start; + u64 end = async_extent->start + async_extent->ram_size - 1; + unsigned long nr_written = 0; + int page_started = 0; + int ret; -again: - while (!list_empty(&async_chunk->extents)) { - async_extent = list_entry(async_chunk->extents.next, - struct async_extent, list); - list_del(&async_extent->list); + /* + * Call cow_file_range() to run the delalloc range directly, since we + * won't go to NOCOW or async path again. + * + * Also we call cow_file_range() with @unlock_page == 0, so that we + * can directly submit them without interruption. + */ + ret = cow_file_range(inode, locked_page, start, end, &page_started, + &nr_written, 0, NULL); + /* Inline extent inserted, page gets unlocked and everything is done */ + if (page_started) { + ret = 0; + goto out; + } + if (ret < 0) { + btrfs_cleanup_ordered_extents(inode, locked_page, start, end - start + 1); + if (locked_page) { + const u64 page_start = page_offset(locked_page); + const u64 page_end = page_start + PAGE_SIZE - 1; -retry: - lock_extent(io_tree, async_extent->start, - async_extent->start + async_extent->ram_size - 1); - /* did the compression code fall back to uncompressed IO? */ - if (!async_extent->pages) { - int page_started = 0; - unsigned long nr_written = 0; + btrfs_page_set_error(inode->root->fs_info, locked_page, + page_start, PAGE_SIZE); + set_page_writeback(locked_page); + end_page_writeback(locked_page); + end_extent_writepage(locked_page, ret, page_start, page_end); + unlock_page(locked_page); + } + goto out; + } - /* allocate blocks */ - ret = cow_file_range(inode, async_chunk->locked_page, - async_extent->start, - async_extent->start + - async_extent->ram_size - 1, - &page_started, &nr_written, 0); + ret = extent_write_locked_range(&inode->vfs_inode, start, end); + /* All pages will be unlocked, including @locked_page */ +out: + kfree(async_extent); + return ret; +} - /* JDM XXX */ +static int submit_one_async_extent(struct btrfs_inode *inode, + struct async_chunk *async_chunk, + struct async_extent *async_extent, + u64 *alloc_hint) +{ + struct extent_io_tree *io_tree = &inode->io_tree; + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_key ins; + struct page *locked_page = NULL; + struct extent_map *em; + int ret = 0; + u64 start = async_extent->start; + u64 end = async_extent->start + async_extent->ram_size - 1; - /* - * if page_started, cow_file_range inserted an - * inline extent and took care of all the unlocking - * and IO for us. Otherwise, we need to submit - * all those pages down to the drive. - */ - if (!page_started && !ret) - extent_write_locked_range(inode, - async_extent->start, - async_extent->start + - async_extent->ram_size - 1, - WB_SYNC_ALL); - else if (ret && async_chunk->locked_page) - unlock_page(async_chunk->locked_page); - kfree(async_extent); - cond_resched(); - continue; - } + /* + * If async_chunk->locked_page is in the async_extent range, we need to + * handle it. + */ + if (async_chunk->locked_page) { + u64 locked_page_start = page_offset(async_chunk->locked_page); + u64 locked_page_end = locked_page_start + PAGE_SIZE - 1; - ret = btrfs_reserve_extent(root, async_extent->ram_size, - async_extent->compressed_size, - async_extent->compressed_size, - 0, alloc_hint, &ins, 1, 1); - if (ret) { - free_async_extent_pages(async_extent); - - if (ret == -ENOSPC) { - unlock_extent(io_tree, async_extent->start, - async_extent->start + - async_extent->ram_size - 1); - - /* - * we need to redirty the pages if we decide to - * fallback to uncompressed IO, otherwise we - * will not submit these pages down to lower - * layers. - */ - extent_range_redirty_for_io(inode, - async_extent->start, - async_extent->start + - async_extent->ram_size - 1); - - goto retry; - } - goto out_free; - } - /* - * here we're doing allocation and writeback of the - * compressed pages - */ - 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); + if (!(start >= locked_page_end || end <= locked_page_start)) + locked_page = async_chunk->locked_page; + } + lock_extent(io_tree, start, end, NULL); - ret = btrfs_add_ordered_extent_compress(inode, - async_extent->start, - ins.objectid, - async_extent->ram_size, - ins.offset, - BTRFS_ORDERED_COMPRESSED, - async_extent->compress_type); - if (ret) { - 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(fs_info, ins.objectid); + /* We have fall back to uncompressed write */ + if (!async_extent->pages) + return submit_uncompressed_range(inode, async_extent, locked_page); + ret = btrfs_reserve_extent(root, async_extent->ram_size, + async_extent->compressed_size, + async_extent->compressed_size, + 0, *alloc_hint, &ins, 1, 1); + if (ret) { + free_async_extent_pages(async_extent); /* - * clear dirty, set writeback and unlock the pages. + * Here we used to try again by going back to non-compressed + * path for ENOSPC. But we can't reserve space even for + * compressed size, how could it work for uncompressed size + * which requires larger size? So here we directly go error + * path. */ - extent_clear_unlock_delalloc(inode, async_extent->start, - async_extent->start + - async_extent->ram_size - 1, - NULL, EXTENT_LOCKED | EXTENT_DELALLOC, - PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK); - if (btrfs_submit_compressed_write(inode, - async_extent->start, - async_extent->ram_size, - ins.objectid, - ins.offset, async_extent->pages, - async_extent->nr_pages, - async_chunk->write_flags, - async_chunk->blkcg_css)) { - struct page *p = async_extent->pages[0]; - const u64 start = async_extent->start; - const u64 end = start + async_extent->ram_size - 1; - - p->mapping = inode->i_mapping; - btrfs_writepage_endio_finish_ordered(p, start, end, 0); - - p->mapping = NULL; - extent_clear_unlock_delalloc(inode, start, end, - NULL, 0, - PAGE_END_WRITEBACK | - PAGE_SET_ERROR); - free_async_extent_pages(async_extent); - } - alloc_hint = ins.objectid + ins.offset; - kfree(async_extent); - cond_resched(); + goto out_free; + } + + /* Here we're doing allocation and writeback of the compressed pages */ + em = create_io_em(inode, start, + async_extent->ram_size, /* len */ + 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 = PTR_ERR(em); + goto out_free_reserve; } - return; + free_extent_map(em); + + ret = btrfs_add_ordered_extent(inode, start, /* file_offset */ + async_extent->ram_size, /* num_bytes */ + async_extent->ram_size, /* ram_bytes */ + ins.objectid, /* disk_bytenr */ + ins.offset, /* disk_num_bytes */ + 0, /* offset */ + 1 << BTRFS_ORDERED_COMPRESSED, + async_extent->compress_type); + if (ret) { + btrfs_drop_extent_map_range(inode, start, end, false); + goto out_free_reserve; + } + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + + /* Clear dirty, set writeback and unlock the pages. */ + extent_clear_unlock_delalloc(inode, start, end, + NULL, EXTENT_LOCKED | EXTENT_DELALLOC, + PAGE_UNLOCK | PAGE_START_WRITEBACK); + if (btrfs_submit_compressed_write(inode, start, /* file_offset */ + async_extent->ram_size, /* num_bytes */ + ins.objectid, /* disk_bytenr */ + ins.offset, /* compressed_len */ + async_extent->pages, /* compressed_pages */ + async_extent->nr_pages, + async_chunk->write_flags, + async_chunk->blkcg_css, true)) { + const u64 start = async_extent->start; + const u64 end = start + async_extent->ram_size - 1; + + btrfs_writepage_endio_finish_ordered(inode, NULL, start, end, 0); + + extent_clear_unlock_delalloc(inode, start, end, NULL, 0, + PAGE_END_WRITEBACK | PAGE_SET_ERROR); + free_async_extent_pages(async_extent); + } + *alloc_hint = ins.objectid + ins.offset; + kfree(async_extent); + return ret; + out_free_reserve: 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 + - async_extent->ram_size - 1, + extent_clear_unlock_delalloc(inode, start, end, NULL, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, - PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK | - PAGE_SET_ERROR); + PAGE_UNLOCK | PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK | PAGE_SET_ERROR); free_async_extent_pages(async_extent); kfree(async_extent); - goto again; + return ret; +} + +/* + * Phase two of compressed writeback. This is the ordered portion of the code, + * which only gets called in the order the work was queued. We walk all the + * async extents created by compress_file_range and send them down to the disk. + */ +static noinline void submit_compressed_extents(struct async_chunk *async_chunk) +{ + struct btrfs_inode *inode = BTRFS_I(async_chunk->inode); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct async_extent *async_extent; + u64 alloc_hint = 0; + int ret = 0; + + while (!list_empty(&async_chunk->extents)) { + u64 extent_start; + u64 ram_size; + + async_extent = list_entry(async_chunk->extents.next, + struct async_extent, list); + list_del(&async_extent->list); + extent_start = async_extent->start; + ram_size = async_extent->ram_size; + + ret = submit_one_async_extent(inode, async_chunk, async_extent, + &alloc_hint); + btrfs_debug(fs_info, +"async extent submission failed root=%lld inode=%llu start=%llu len=%llu ret=%d", + inode->root->root_key.objectid, + btrfs_ino(inode), extent_start, ram_size, ret); + } } -static u64 get_extent_allocation_hint(struct inode *inode, u64 start, +static u64 get_extent_allocation_hint(struct btrfs_inode *inode, u64 start, u64 num_bytes) { - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; u64 alloc_hint = 0; @@ -962,18 +1145,43 @@ static u64 get_extent_allocation_hint(struct inode *inode, u64 start, * *page_started is set to one if we unlock locked_page and do everything * required to start IO on it. It may be clean and already done with * IO when we return. + * + * When unlock == 1, we unlock the pages in successfully allocated regions. + * When unlock == 0, we leave them locked for writing them out. + * + * However, we unlock all the pages except @locked_page in case of failure. + * + * In summary, page locking state will be as follow: + * + * - page_started == 1 (return value) + * - All the pages are unlocked. IO is started. + * - Note that this can happen only on success + * - unlock == 1 + * - All the pages except @locked_page are unlocked in any case + * - unlock == 0 + * - On success, all the pages are locked for writing out them + * - On failure, all the pages except @locked_page are unlocked + * + * When a failure happens in the second or later iteration of the + * while-loop, the ordered extents created in previous iterations are kept + * intact. So, the caller must clean them up by calling + * btrfs_cleanup_ordered_extents(). See btrfs_run_delalloc_range() for + * example. */ -static noinline int cow_file_range(struct inode *inode, +static noinline int cow_file_range(struct btrfs_inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, - unsigned long *nr_written, int unlock) + unsigned long *nr_written, int unlock, + u64 *done_offset) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; u64 alloc_hint = 0; + u64 orig_start = start; u64 num_bytes; unsigned long ram_size; u64 cur_alloc_size = 0; + u64 min_alloc_size; u64 blocksize = fs_info->sectorsize; struct btrfs_key ins; struct extent_map *em; @@ -982,8 +1190,7 @@ static noinline int cow_file_range(struct inode *inode, bool extent_reserved = false; int ret = 0; - if (btrfs_is_free_space_inode(BTRFS_I(inode))) { - WARN_ON_ONCE(1); + if (btrfs_is_free_space_inode(inode)) { ret = -EINVAL; goto out_unlock; } @@ -992,12 +1199,25 @@ static noinline int cow_file_range(struct inode *inode, num_bytes = max(blocksize, num_bytes); ASSERT(num_bytes <= btrfs_super_total_bytes(fs_info->super_copy)); - inode_should_defrag(BTRFS_I(inode), start, end, num_bytes, SZ_64K); + inode_should_defrag(inode, start, end, num_bytes, SZ_64K); + + /* + * Due to the page size limit, for subpage we can only trigger the + * writeback for the dirty sectors of page, that means data writeback + * is doing more writeback than what we want. + * + * This is especially unexpected for some call sites like fallocate, + * where we only increase i_size after everything is done. + * This means we can trigger inline extent even if we didn't want to. + * So here we skip inline extent creation completely. + */ + if (start == 0 && fs_info->sectorsize == PAGE_SIZE) { + u64 actual_end = min_t(u64, i_size_read(&inode->vfs_inode), + end + 1); - if (start == 0) { /* lets try to make an inline extent */ - ret = cow_file_range_inline(inode, start, end, 0, - BTRFS_COMPRESS_NONE, NULL); + ret = cow_file_range_inline(inode, actual_end, 0, + BTRFS_COMPRESS_NONE, NULL, false); if (ret == 0) { /* * We use DO_ACCOUNTING here because we need the @@ -1005,15 +1225,28 @@ static noinline int cow_file_range(struct inode *inode, * our outstanding extent for clearing delalloc for this * range. */ - extent_clear_unlock_delalloc(inode, start, end, NULL, + extent_clear_unlock_delalloc(inode, start, end, + locked_page, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK | - PAGE_END_WRITEBACK); + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); *nr_written = *nr_written + (end - start + PAGE_SIZE) / PAGE_SIZE; *page_started = 1; + /* + * locked_page is locked by the caller of + * writepage_delalloc(), not locked by + * __process_pages_contig(). + * + * We can't let __process_pages_contig() to unlock it, + * as it doesn't have any subpage::writers recorded. + * + * Here we manually unlock the page, since the caller + * can't use page_started to determine if it's an + * inline extent or a compressed extent. + */ + unlock_page(locked_page); goto out; } else if (ret < 0) { goto out_unlock; @@ -1021,13 +1254,27 @@ static noinline int cow_file_range(struct inode *inode, } alloc_hint = get_extent_allocation_hint(inode, start, num_bytes); - btrfs_drop_extent_cache(BTRFS_I(inode), start, - start + num_bytes - 1, 0); + + /* + * Relocation relies on the relocated extents to have exactly the same + * size as the original extents. Normally writeback for relocation data + * extents follows a NOCOW path because relocation preallocates the + * extents. However, due to an operation such as scrub turning a block + * group to RO mode, it may fallback to COW mode, so we must make sure + * an extent allocated during COW has exactly the requested size and can + * not be split into smaller extents, otherwise relocation breaks and + * fails during the stage where it updates the bytenr of file extent + * items. + */ + if (btrfs_is_data_reloc_root(root)) + min_alloc_size = num_bytes; + else + min_alloc_size = fs_info->sectorsize; while (num_bytes > 0) { cur_alloc_size = num_bytes; ret = btrfs_reserve_extent(root, cur_alloc_size, cur_alloc_size, - fs_info->sectorsize, 0, alloc_hint, + min_alloc_size, 0, alloc_hint, &ins, 1, 1); if (ret < 0) goto out_unlock; @@ -1049,13 +1296,14 @@ static noinline int cow_file_range(struct inode *inode, } free_extent_map(em); - ret = btrfs_add_ordered_extent(inode, start, ins.objectid, - ram_size, cur_alloc_size, 0); + ret = btrfs_add_ordered_extent(inode, start, ram_size, ram_size, + ins.objectid, cur_alloc_size, 0, + 1 << BTRFS_ORDERED_REGULAR, + BTRFS_COMPRESS_NONE); if (ret) goto out_drop_extent_cache; - if (root->root_key.objectid == - BTRFS_DATA_RELOC_TREE_OBJECTID) { + if (btrfs_is_data_reloc_root(root)) { ret = btrfs_reloc_clone_csums(inode, start, cur_alloc_size); /* @@ -1070,24 +1318,25 @@ static noinline int cow_file_range(struct inode *inode, * skip current ordered extent. */ if (ret) - btrfs_drop_extent_cache(BTRFS_I(inode), start, - start + ram_size - 1, 0); + btrfs_drop_extent_map_range(inode, start, + start + ram_size - 1, + false); } btrfs_dec_block_group_reservations(fs_info, ins.objectid); - /* we're not doing compressed IO, don't unlock the first - * page (which the caller expects to stay locked), don't - * clear any dirty bits and don't set any writeback bits + /* + * We're not doing compressed IO, don't unlock the first page + * (which the caller expects to stay locked), don't clear any + * dirty bits and don't set any writeback bits * - * Do set the Private2 bit so we know this page was properly - * setup for writepage + * Do set the Ordered (Private2) bit so we know this page was + * properly setup for writepage. */ page_ops = unlock ? PAGE_UNLOCK : 0; - page_ops |= PAGE_SET_PRIVATE2; + page_ops |= PAGE_SET_ORDERED; - extent_clear_unlock_delalloc(inode, start, - start + ram_size - 1, + extent_clear_unlock_delalloc(inode, start, start + ram_size - 1, locked_page, EXTENT_LOCKED | EXTENT_DELALLOC, page_ops); @@ -1111,39 +1360,89 @@ out: return ret; out_drop_extent_cache: - btrfs_drop_extent_cache(BTRFS_I(inode), start, start + ram_size - 1, 0); + btrfs_drop_extent_map_range(inode, start, start + ram_size - 1, false); out_reserve: btrfs_dec_block_group_reservations(fs_info, ins.objectid); btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); out_unlock: + /* + * If done_offset is non-NULL and ret == -EAGAIN, we expect the + * caller to write out the successfully allocated region and retry. + */ + if (done_offset && ret == -EAGAIN) { + if (orig_start < start) + *done_offset = start - 1; + else + *done_offset = start; + return ret; + } else if (ret == -EAGAIN) { + /* Convert to -ENOSPC since the caller cannot retry. */ + ret = -ENOSPC; + } + + /* + * Now, we have three regions to clean up: + * + * |-------(1)----|---(2)---|-------------(3)----------| + * `- orig_start `- start `- start + cur_alloc_size `- end + * + * We process each region below. + */ + clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV; - page_ops = PAGE_UNLOCK | PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK | - PAGE_END_WRITEBACK; - /* - * If we reserved an extent for our delalloc range (or a subrange) and - * failed to create the respective ordered extent, then it means that - * when we reserved the extent we decremented the extent's size from - * the data space_info's bytes_may_use counter and incremented the - * space_info's bytes_reserved counter by the same amount. We must make - * sure extent_clear_unlock_delalloc() does not try to decrement again - * the data space_info's bytes_may_use counter, therefore we do not pass - * it the flag EXTENT_CLEAR_DATA_RESV. + page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK; + + /* + * For the range (1). We have already instantiated the ordered extents + * for this region. They are cleaned up by + * btrfs_cleanup_ordered_extents() in e.g, + * btrfs_run_delalloc_range(). EXTENT_LOCKED | EXTENT_DELALLOC are + * already cleared in the above loop. And, EXTENT_DELALLOC_NEW | + * EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV are handled by the cleanup + * function. + * + * However, in case of unlock == 0, we still need to unlock the pages + * (except @locked_page) to ensure all the pages are unlocked. + */ + if (!unlock && orig_start < start) { + if (!locked_page) + mapping_set_error(inode->vfs_inode.i_mapping, ret); + extent_clear_unlock_delalloc(inode, orig_start, start - 1, + locked_page, 0, page_ops); + } + + /* + * For the range (2). If we reserved an extent for our delalloc range + * (or a subrange) and failed to create the respective ordered extent, + * then it means that when we reserved the extent we decremented the + * extent's size from the data space_info's bytes_may_use counter and + * incremented the space_info's bytes_reserved counter by the same + * amount. We must make sure extent_clear_unlock_delalloc() does not try + * to decrement again the data space_info's bytes_may_use counter, + * therefore we do not pass it the flag EXTENT_CLEAR_DATA_RESV. */ if (extent_reserved) { extent_clear_unlock_delalloc(inode, start, - start + cur_alloc_size, + start + cur_alloc_size - 1, locked_page, clear_bits, page_ops); start += cur_alloc_size; if (start >= end) - goto out; + return ret; } + + /* + * For the range (3). We never touched the region. In addition to the + * clear_bits above, we add EXTENT_CLEAR_DATA_RESV to release the data + * space_info's bytes_may_use counter, reserved in + * btrfs_check_data_free_space(). + */ extent_clear_unlock_delalloc(inode, start, end, locked_page, clear_bits | EXTENT_CLEAR_DATA_RESV, page_ops); - goto out; + return ret; } /* @@ -1176,11 +1475,6 @@ static noinline void async_cow_submit(struct btrfs_work *work) nr_pages = (async_chunk->end - async_chunk->start + PAGE_SIZE) >> PAGE_SHIFT; - /* atomic_sub_return implies a barrier */ - if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) < - 5 * SZ_1M) - cond_wake_up_nomb(&fs_info->async_submit_wait); - /* * ->inode could be NULL if async_chunk_start has failed to compress, * in which case we don't have anything to submit, yet we need to @@ -1189,32 +1483,36 @@ static noinline void async_cow_submit(struct btrfs_work *work) */ if (async_chunk->inode) submit_compressed_extents(async_chunk); + + /* atomic_sub_return implies a barrier */ + if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) < + 5 * SZ_1M) + cond_wake_up_nomb(&fs_info->async_submit_wait); } static noinline void async_cow_free(struct btrfs_work *work) { struct async_chunk *async_chunk; + struct async_cow *async_cow; async_chunk = container_of(work, struct async_chunk, work); if (async_chunk->inode) btrfs_add_delayed_iput(async_chunk->inode); if (async_chunk->blkcg_css) css_put(async_chunk->blkcg_css); - /* - * Since the pointer to 'pending' is at the beginning of the array of - * async_chunk's, freeing it ensures the whole array has been freed. - */ - if (atomic_dec_and_test(async_chunk->pending)) - kvfree(async_chunk->pending); + + async_cow = async_chunk->async_cow; + if (atomic_dec_and_test(&async_cow->num_chunks)) + kvfree(async_cow); } -static int cow_file_range_async(struct inode *inode, +static int cow_file_range_async(struct btrfs_inode *inode, struct writeback_control *wbc, 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 btrfs_fs_info *fs_info = inode->root->fs_info; struct cgroup_subsys_state *blkcg_css = wbc_blkcg_css(wbc); struct async_cow *ctx; struct async_chunk *async_chunk; @@ -1224,11 +1522,11 @@ static int cow_file_range_async(struct inode *inode, int i; bool should_compress; unsigned nofs_flag; - const unsigned int write_flags = wbc_to_write_flags(wbc); + const blk_opf_t write_flags = wbc_to_write_flags(wbc); - unlock_extent(&BTRFS_I(inode)->io_tree, start, end); + unlock_extent(&inode->io_tree, start, end, NULL); - if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS && + if (inode->flags & BTRFS_INODE_NOCOMPRESS && !btrfs_test_opt(fs_info, FORCE_COMPRESS)) { num_chunks = 1; should_compress = false; @@ -1244,9 +1542,8 @@ static int cow_file_range_async(struct inode *inode, unsigned clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING; - unsigned long page_ops = PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK | - PAGE_SET_ERROR; + unsigned long page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK | PAGE_SET_ERROR; extent_clear_unlock_delalloc(inode, start, end, locked_page, clear_bits, page_ops); @@ -1266,9 +1563,9 @@ static int cow_file_range_async(struct inode *inode, * igrab is called higher up in the call chain, take only the * lightweight reference for the callback lifetime */ - ihold(inode); - async_chunk[i].pending = &ctx->num_chunks; - async_chunk[i].inode = inode; + ihold(&inode->vfs_inode); + async_chunk[i].async_cow = ctx; + async_chunk[i].inode = &inode->vfs_inode; async_chunk[i].start = start; async_chunk[i].end = cur_end; async_chunk[i].write_flags = write_flags; @@ -1323,15 +1620,62 @@ static int cow_file_range_async(struct inode *inode, return 0; } -static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, - u64 bytenr, u64 num_bytes) +static noinline int run_delalloc_zoned(struct btrfs_inode *inode, + struct page *locked_page, u64 start, + u64 end, int *page_started, + unsigned long *nr_written) { + u64 done_offset = end; int ret; + bool locked_page_done = false; + + while (start <= end) { + ret = cow_file_range(inode, locked_page, start, end, page_started, + nr_written, 0, &done_offset); + if (ret && ret != -EAGAIN) + return ret; + + if (*page_started) { + ASSERT(ret == 0); + return 0; + } + + if (ret == 0) + done_offset = end; + + if (done_offset == start) { + wait_on_bit_io(&inode->root->fs_info->flags, + BTRFS_FS_NEED_ZONE_FINISH, + TASK_UNINTERRUPTIBLE); + continue; + } + + if (!locked_page_done) { + __set_page_dirty_nobuffers(locked_page); + account_page_redirty(locked_page); + } + locked_page_done = true; + extent_write_locked_range(&inode->vfs_inode, start, done_offset); + + start = done_offset + 1; + } + + *page_started = 1; + + return 0; +} + +static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, + u64 bytenr, u64 num_bytes, bool nowait) +{ + struct btrfs_root *csum_root = btrfs_csum_root(fs_info, bytenr); struct btrfs_ordered_sum *sums; + int ret; LIST_HEAD(list); - ret = btrfs_lookup_csums_range(fs_info->csum_root, bytenr, - bytenr + num_bytes - 1, &list, 0); + ret = btrfs_lookup_csums_range(csum_root, bytenr, + bytenr + num_bytes - 1, &list, 0, + nowait); if (ret == 0 && list_empty(&list)) return 0; @@ -1345,6 +1689,209 @@ static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, return 1; } +static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, + const u64 start, const u64 end, + int *page_started, unsigned long *nr_written) +{ + const bool is_space_ino = btrfs_is_free_space_inode(inode); + const bool is_reloc_ino = btrfs_is_data_reloc_root(inode->root); + const u64 range_bytes = end + 1 - start; + struct extent_io_tree *io_tree = &inode->io_tree; + u64 range_start = start; + u64 count; + + /* + * If EXTENT_NORESERVE is set it means that when the buffered write was + * made we had not enough available data space and therefore we did not + * reserve data space for it, since we though we could do NOCOW for the + * respective file range (either there is prealloc extent or the inode + * has the NOCOW bit set). + * + * However when we need to fallback to COW mode (because for example the + * block group for the corresponding extent was turned to RO mode by a + * scrub or relocation) we need to do the following: + * + * 1) We increment the bytes_may_use counter of the data space info. + * If COW succeeds, it allocates a new data extent and after doing + * that it decrements the space info's bytes_may_use counter and + * increments its bytes_reserved counter by the same amount (we do + * this at btrfs_add_reserved_bytes()). So we need to increment the + * bytes_may_use counter to compensate (when space is reserved at + * buffered write time, the bytes_may_use counter is incremented); + * + * 2) We clear the EXTENT_NORESERVE bit from the range. We do this so + * that if the COW path fails for any reason, it decrements (through + * extent_clear_unlock_delalloc()) the bytes_may_use counter of the + * data space info, which we incremented in the step above. + * + * If we need to fallback to cow and the inode corresponds to a free + * space cache inode or an inode of the data relocation tree, we must + * also increment bytes_may_use of the data space_info for the same + * reason. Space caches and relocated data extents always get a prealloc + * extent for them, however scrub or balance may have set the block + * group that contains that extent to RO mode and therefore force COW + * when starting writeback. + */ + count = count_range_bits(io_tree, &range_start, end, range_bytes, + EXTENT_NORESERVE, 0); + if (count > 0 || is_space_ino || is_reloc_ino) { + u64 bytes = count; + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct btrfs_space_info *sinfo = fs_info->data_sinfo; + + if (is_space_ino || is_reloc_ino) + bytes = range_bytes; + + spin_lock(&sinfo->lock); + btrfs_space_info_update_bytes_may_use(fs_info, sinfo, bytes); + spin_unlock(&sinfo->lock); + + if (count > 0) + clear_extent_bit(io_tree, start, end, EXTENT_NORESERVE, + NULL); + } + + return cow_file_range(inode, locked_page, start, end, page_started, + nr_written, 1, NULL); +} + +struct can_nocow_file_extent_args { + /* Input fields. */ + + /* Start file offset of the range we want to NOCOW. */ + u64 start; + /* End file offset (inclusive) of the range we want to NOCOW. */ + u64 end; + bool writeback_path; + bool strict; + /* + * Free the path passed to can_nocow_file_extent() once it's not needed + * anymore. + */ + bool free_path; + + /* Output fields. Only set when can_nocow_file_extent() returns 1. */ + + u64 disk_bytenr; + u64 disk_num_bytes; + u64 extent_offset; + /* Number of bytes that can be written to in NOCOW mode. */ + u64 num_bytes; +}; + +/* + * Check if we can NOCOW the file extent that the path points to. + * This function may return with the path released, so the caller should check + * if path->nodes[0] is NULL or not if it needs to use the path afterwards. + * + * Returns: < 0 on error + * 0 if we can not NOCOW + * 1 if we can NOCOW + */ +static int can_nocow_file_extent(struct btrfs_path *path, + struct btrfs_key *key, + struct btrfs_inode *inode, + struct can_nocow_file_extent_args *args) +{ + const bool is_freespace_inode = btrfs_is_free_space_inode(inode); + struct extent_buffer *leaf = path->nodes[0]; + struct btrfs_root *root = inode->root; + struct btrfs_file_extent_item *fi; + u64 extent_end; + u8 extent_type; + int can_nocow = 0; + int ret = 0; + bool nowait = path->nowait; + + fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + extent_type = btrfs_file_extent_type(leaf, fi); + + if (extent_type == BTRFS_FILE_EXTENT_INLINE) + goto out; + + /* Can't access these fields unless we know it's not an inline extent. */ + args->disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); + args->disk_num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi); + args->extent_offset = btrfs_file_extent_offset(leaf, fi); + + if (!(inode->flags & BTRFS_INODE_NODATACOW) && + extent_type == BTRFS_FILE_EXTENT_REG) + goto out; + + /* + * If the extent was created before the generation where the last snapshot + * for its subvolume was created, then this implies the extent is shared, + * hence we must COW. + */ + if (!args->strict && + btrfs_file_extent_generation(leaf, fi) <= + btrfs_root_last_snapshot(&root->root_item)) + goto out; + + /* An explicit hole, must COW. */ + if (args->disk_bytenr == 0) + goto out; + + /* Compressed/encrypted/encoded extents must be COWed. */ + if (btrfs_file_extent_compression(leaf, fi) || + btrfs_file_extent_encryption(leaf, fi) || + btrfs_file_extent_other_encoding(leaf, fi)) + goto out; + + extent_end = btrfs_file_extent_end(path); + + /* + * The following checks can be expensive, as they need to take other + * locks and do btree or rbtree searches, so release the path to avoid + * blocking other tasks for too long. + */ + btrfs_release_path(path); + + ret = btrfs_cross_ref_exist(root, btrfs_ino(inode), + key->offset - args->extent_offset, + args->disk_bytenr, false, path); + WARN_ON_ONCE(ret > 0 && is_freespace_inode); + if (ret != 0) + goto out; + + if (args->free_path) { + /* + * We don't need the path anymore, plus through the + * csum_exist_in_range() call below we will end up allocating + * another path. So free the path to avoid unnecessary extra + * memory usage. + */ + btrfs_free_path(path); + path = NULL; + } + + /* If there are pending snapshots for this root, we must COW. */ + if (args->writeback_path && !is_freespace_inode && + atomic_read(&root->snapshot_force_cow)) + goto out; + + args->disk_bytenr += args->extent_offset; + args->disk_bytenr += args->start - key->offset; + args->num_bytes = min(args->end + 1, extent_end) - args->start; + + /* + * Force COW if csums exist in the range. This ensures that csums for a + * given extent are either valid or do not exist. + */ + ret = csum_exist_in_range(root->fs_info, args->disk_bytenr, args->num_bytes, + nowait); + WARN_ON_ONCE(ret > 0 && is_freespace_inode); + if (ret != 0) + goto out; + + can_nocow = 1; + out: + if (args->free_path && path) + btrfs_free_path(path); + + return ret < 0 ? ret : can_nocow; +} + /* * when nowcow writeback call back. This checks for snapshots or COW copies * of the extents that exist in the file, and COWs the file as required. @@ -1352,23 +1899,23 @@ static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, * If no cow copies or snapshots exist, we write directly to the existing * blocks on disk */ -static noinline int run_delalloc_nocow(struct inode *inode, +static noinline int run_delalloc_nocow(struct btrfs_inode *inode, struct page *locked_page, const u64 start, const u64 end, - int *page_started, int force, + int *page_started, 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_fs_info *fs_info = inode->root->fs_info; + struct btrfs_root *root = inode->root; struct btrfs_path *path; u64 cow_start = (u64)-1; u64 cur_offset = start; int ret; bool check_prev = true; - const bool freespace_inode = btrfs_is_free_space_inode(BTRFS_I(inode)); - u64 ino = btrfs_ino(BTRFS_I(inode)); + u64 ino = btrfs_ino(inode); + struct btrfs_block_group *bg; bool nocow = false; - u64 disk_bytenr = 0; + struct can_nocow_file_extent_args nocow_args = { 0 }; path = btrfs_alloc_path(); if (!path) { @@ -1376,21 +1923,21 @@ static noinline int run_delalloc_nocow(struct inode *inode, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); return -ENOMEM; } + nocow_args.end = end; + nocow_args.writeback_path = true; + while (1) { struct btrfs_key found_key; struct btrfs_file_extent_item *fi; struct extent_buffer *leaf; u64 extent_end; - u64 extent_offset; - u64 num_bytes = 0; - u64 disk_num_bytes; u64 ram_bytes; + u64 nocow_end; int extent_type; nocow = false; @@ -1466,109 +2013,38 @@ next_slot: fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); extent_type = btrfs_file_extent_type(leaf, fi); - + /* If this is triggered then we have a memory corruption. */ + ASSERT(extent_type < BTRFS_NR_FILE_EXTENT_TYPES); + if (WARN_ON(extent_type >= BTRFS_NR_FILE_EXTENT_TYPES)) { + ret = -EUCLEAN; + goto error; + } ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); - if (extent_type == BTRFS_FILE_EXTENT_REG || - extent_type == BTRFS_FILE_EXTENT_PREALLOC) { - disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); - extent_offset = btrfs_file_extent_offset(leaf, fi); - extent_end = found_key.offset + - btrfs_file_extent_num_bytes(leaf, fi); - disk_num_bytes = - btrfs_file_extent_disk_num_bytes(leaf, fi); - /* - * If the extent we got ends before our current offset, - * skip to the next extent. - */ - if (extent_end <= cur_offset) { - path->slots[0]++; - goto next_slot; - } - /* Skip holes */ - if (disk_bytenr == 0) - goto out_check; - /* Skip compressed/encrypted/encoded extents */ - if (btrfs_file_extent_compression(leaf, fi) || - btrfs_file_extent_encryption(leaf, fi) || - btrfs_file_extent_other_encoding(leaf, fi)) - goto out_check; - /* - * If extent is created before the last volume's snapshot - * this implies the extent is shared, hence we can't do - * nocow. This is the same check as in - * btrfs_cross_ref_exist but without calling - * btrfs_search_slot. - */ - if (!freespace_inode && - btrfs_file_extent_generation(leaf, fi) <= - btrfs_root_last_snapshot(&root->root_item)) - goto out_check; - if (extent_type == BTRFS_FILE_EXTENT_REG && !force) - goto out_check; - /* If extent is RO, we must COW it */ - if (btrfs_extent_readonly(fs_info, disk_bytenr)) - goto out_check; - ret = btrfs_cross_ref_exist(root, ino, - found_key.offset - - extent_offset, disk_bytenr); - if (ret) { - /* - * ret could be -EIO if the above fails to read - * metadata. - */ - if (ret < 0) { - if (cow_start != (u64)-1) - cur_offset = cow_start; - goto error; - } + extent_end = btrfs_file_extent_end(path); - WARN_ON_ONCE(freespace_inode); - goto out_check; - } - disk_bytenr += extent_offset; - disk_bytenr += cur_offset - found_key.offset; - num_bytes = min(end + 1, extent_end) - cur_offset; - /* - * If there are pending snapshots for this root, we - * fall into common COW way - */ - if (!freespace_inode && atomic_read(&root->snapshot_force_cow)) - goto out_check; - /* - * force cow if csum exists in the range. - * this ensure that csum for a given extent are - * either valid or do not exist. - */ - ret = csum_exist_in_range(fs_info, disk_bytenr, - num_bytes); - if (ret) { - /* - * ret could be -EIO if the above fails to read - * metadata. - */ - if (ret < 0) { - if (cow_start != (u64)-1) - cur_offset = cow_start; - goto error; - } - WARN_ON_ONCE(freespace_inode); - goto out_check; - } - if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) - goto out_check; - nocow = true; - } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - extent_end = found_key.offset + ram_bytes; - extent_end = ALIGN(extent_end, fs_info->sectorsize); - /* Skip extents outside of our requested range */ - if (extent_end <= start) { - path->slots[0]++; - goto next_slot; - } - } else { - /* If this triggers then we have a memory corruption */ - BUG(); + /* + * If the extent we got ends before our current offset, skip to + * the next extent. + */ + if (extent_end <= cur_offset) { + path->slots[0]++; + goto next_slot; + } + + nocow_args.start = cur_offset; + ret = can_nocow_file_extent(path, &found_key, inode, &nocow_args); + if (ret < 0) { + if (cow_start != (u64)-1) + cur_offset = cow_start; + goto error; + } else if (ret == 0) { + goto out_check; } + + ret = 0; + bg = btrfs_inc_nocow_writers(fs_info, nocow_args.disk_bytenr); + if (bg) + nocow = true; out_check: /* * If nocow is false then record the beginning of the range @@ -1580,89 +2056,88 @@ out_check: cur_offset = extent_end; if (cur_offset > end) break; + if (!path->nodes[0]) + continue; path->slots[0]++; goto next_slot; } - btrfs_release_path(path); - /* * COW range from cow_start to found_key.offset - 1. As the key * will contain the beginning of the first extent that can be * NOCOW, following one which needs to be COW'ed */ if (cow_start != (u64)-1) { - ret = cow_file_range(inode, locked_page, - cow_start, found_key.offset - 1, - page_started, nr_written, 1); - if (ret) { - if (nocow) - btrfs_dec_nocow_writers(fs_info, - disk_bytenr); + ret = fallback_to_cow(inode, locked_page, + cow_start, found_key.offset - 1, + page_started, nr_written); + if (ret) goto error; - } cow_start = (u64)-1; } + nocow_end = cur_offset + nocow_args.num_bytes - 1; + if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) { - u64 orig_start = found_key.offset - extent_offset; + u64 orig_start = found_key.offset - nocow_args.extent_offset; struct extent_map *em; - em = create_io_em(inode, cur_offset, num_bytes, + em = create_io_em(inode, cur_offset, nocow_args.num_bytes, orig_start, - disk_bytenr, /* block_start */ - num_bytes, /* block_len */ - disk_num_bytes, /* orig_block_len */ + nocow_args.disk_bytenr, /* block_start */ + nocow_args.num_bytes, /* block_len */ + nocow_args.disk_num_bytes, /* orig_block_len */ ram_bytes, BTRFS_COMPRESS_NONE, BTRFS_ORDERED_PREALLOC); if (IS_ERR(em)) { - if (nocow) - btrfs_dec_nocow_writers(fs_info, - disk_bytenr); ret = PTR_ERR(em); goto error; } free_extent_map(em); - ret = btrfs_add_ordered_extent(inode, cur_offset, - disk_bytenr, num_bytes, - num_bytes, - BTRFS_ORDERED_PREALLOC); + ret = btrfs_add_ordered_extent(inode, + cur_offset, nocow_args.num_bytes, + nocow_args.num_bytes, + nocow_args.disk_bytenr, + nocow_args.num_bytes, 0, + 1 << BTRFS_ORDERED_PREALLOC, + BTRFS_COMPRESS_NONE); if (ret) { - btrfs_drop_extent_cache(BTRFS_I(inode), - cur_offset, - cur_offset + num_bytes - 1, - 0); + btrfs_drop_extent_map_range(inode, cur_offset, + nocow_end, false); goto error; } } else { ret = btrfs_add_ordered_extent(inode, cur_offset, - disk_bytenr, num_bytes, - num_bytes, - BTRFS_ORDERED_NOCOW); + nocow_args.num_bytes, + nocow_args.num_bytes, + nocow_args.disk_bytenr, + nocow_args.num_bytes, + 0, + 1 << BTRFS_ORDERED_NOCOW, + BTRFS_COMPRESS_NONE); if (ret) goto error; } - if (nocow) - btrfs_dec_nocow_writers(fs_info, disk_bytenr); - nocow = false; + if (nocow) { + btrfs_dec_nocow_writers(bg); + nocow = false; + } - if (root->root_key.objectid == - BTRFS_DATA_RELOC_TREE_OBJECTID) + if (btrfs_is_data_reloc_root(root)) /* * Error handled later, as we must prevent * extent_clear_unlock_delalloc() in error handler * from freeing metadata of created ordered extent. */ ret = btrfs_reloc_clone_csums(inode, cur_offset, - num_bytes); + nocow_args.num_bytes); - extent_clear_unlock_delalloc(inode, cur_offset, - cur_offset + num_bytes - 1, + extent_clear_unlock_delalloc(inode, cur_offset, nocow_end, locked_page, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_CLEAR_DATA_RESV, - PAGE_UNLOCK | PAGE_SET_PRIVATE2); + PAGE_UNLOCK | PAGE_SET_ORDERED); cur_offset = extent_end; @@ -1683,75 +2158,82 @@ out_check: if (cow_start != (u64)-1) { cur_offset = end; - ret = cow_file_range(inode, locked_page, cow_start, end, - page_started, nr_written, 1); + ret = fallback_to_cow(inode, locked_page, cow_start, end, + page_started, nr_written); if (ret) goto error; } error: if (nocow) - btrfs_dec_nocow_writers(fs_info, disk_bytenr); + btrfs_dec_nocow_writers(bg); if (ret && cur_offset < end) extent_clear_unlock_delalloc(inode, cur_offset, end, locked_page, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); btrfs_free_path(path); return ret; } -static inline int need_force_cow(struct inode *inode, u64 start, u64 end) +static bool should_nocow(struct btrfs_inode *inode, u64 start, u64 end) { - - if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) && - !(BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC)) - return 0; - - /* - * @defrag_bytes is a hint value, no spinlock held here, - * if is not zero, it means the file is defragging. - * Force cow if given extent needs to be defragged. - */ - if (BTRFS_I(inode)->defrag_bytes && - test_range_bit(&BTRFS_I(inode)->io_tree, start, end, - EXTENT_DEFRAG, 0, NULL)) - return 1; - - return 0; + if (inode->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)) { + if (inode->defrag_bytes && + test_range_bit(&inode->io_tree, start, end, EXTENT_DEFRAG, + 0, NULL)) + return false; + return true; + } + return false; } /* * Function to process delayed allocation (create CoW) for ranges which are * being touched for the first time. */ -int btrfs_run_delalloc_range(struct inode *inode, struct page *locked_page, +int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written, struct writeback_control *wbc) { int ret; - int force_cow = need_force_cow(inode, start, end); + const bool zoned = btrfs_is_zoned(inode->root->fs_info); - if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW && !force_cow) { - ret = run_delalloc_nocow(inode, locked_page, start, end, - page_started, 1, nr_written); - } else if (BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC && !force_cow) { + /* + * The range must cover part of the @locked_page, or the returned + * @page_started can confuse the caller. + */ + ASSERT(!(end <= page_offset(locked_page) || + start >= page_offset(locked_page) + PAGE_SIZE)); + + if (should_nocow(inode, start, end)) { + /* + * Normally on a zoned device we're only doing COW writes, but + * in case of relocation on a zoned filesystem we have taken + * precaution, that we're only writing sequentially. It's safe + * to use run_delalloc_nocow() here, like for regular + * preallocated inodes. + */ + ASSERT(!zoned || btrfs_is_data_reloc_root(inode->root)); ret = run_delalloc_nocow(inode, locked_page, start, end, - page_started, 0, nr_written); - } else if (!inode_can_compress(inode) || + page_started, nr_written); + } else if (!btrfs_inode_can_compress(inode) || !inode_need_compress(inode, start, end)) { - ret = cow_file_range(inode, locked_page, start, end, - page_started, nr_written, 1); + if (zoned) + ret = run_delalloc_zoned(inode, locked_page, start, end, + page_started, nr_written); + else + ret = cow_file_range(inode, locked_page, start, end, + page_started, nr_written, 1, NULL); } else { - set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, - &BTRFS_I(inode)->runtime_flags); + set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, &inode->runtime_flags); ret = cow_file_range_async(inode, wbc, locked_page, start, end, page_started, nr_written); } + ASSERT(ret <= 0); if (ret) btrfs_cleanup_ordered_extents(inode, locked_page, start, end - start + 1); @@ -1761,6 +2243,7 @@ int btrfs_run_delalloc_range(struct inode *inode, struct page *locked_page, void btrfs_split_delalloc_extent(struct inode *inode, struct extent_state *orig, u64 split) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 size; /* not delalloc, ignore it */ @@ -1768,7 +2251,7 @@ void btrfs_split_delalloc_extent(struct inode *inode, return; size = orig->end - orig->start + 1; - if (size > BTRFS_MAX_EXTENT_SIZE) { + if (size > fs_info->max_extent_size) { u32 num_extents; u64 new_size; @@ -1777,10 +2260,10 @@ void btrfs_split_delalloc_extent(struct inode *inode, * applies here, just in reverse. */ new_size = orig->end - split + 1; - num_extents = count_max_extents(new_size); + num_extents = count_max_extents(fs_info, new_size); new_size = split - orig->start; - num_extents += count_max_extents(new_size); - if (count_max_extents(size) >= num_extents) + num_extents += count_max_extents(fs_info, new_size); + if (count_max_extents(fs_info, size) >= num_extents) return; } @@ -1797,6 +2280,7 @@ void btrfs_split_delalloc_extent(struct inode *inode, void btrfs_merge_delalloc_extent(struct inode *inode, struct extent_state *new, struct extent_state *other) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 new_size, old_size; u32 num_extents; @@ -1810,7 +2294,7 @@ void btrfs_merge_delalloc_extent(struct inode *inode, struct extent_state *new, new_size = other->end - new->start + 1; /* we're not bigger than the max, unreserve the space and go */ - if (new_size <= BTRFS_MAX_EXTENT_SIZE) { + if (new_size <= fs_info->max_extent_size) { spin_lock(&BTRFS_I(inode)->lock); btrfs_mod_outstanding_extents(BTRFS_I(inode), -1); spin_unlock(&BTRFS_I(inode)->lock); @@ -1836,10 +2320,10 @@ void btrfs_merge_delalloc_extent(struct inode *inode, struct extent_state *new, * this case. */ old_size = other->end - other->start + 1; - num_extents = count_max_extents(old_size); + num_extents = count_max_extents(fs_info, old_size); old_size = new->end - new->start + 1; - num_extents += count_max_extents(old_size); - if (count_max_extents(new_size) >= num_extents) + num_extents += count_max_extents(fs_info, old_size); + if (count_max_extents(fs_info, new_size) >= num_extents) return; spin_lock(&BTRFS_I(inode)->lock); @@ -1904,21 +2388,21 @@ static void btrfs_del_delalloc_inode(struct btrfs_root *root, * list of inodes that have pending delalloc work to be done. */ void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state, - unsigned *bits) + u32 bits) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - if ((*bits & EXTENT_DEFRAG) && !(*bits & EXTENT_DELALLOC)) + if ((bits & EXTENT_DEFRAG) && !(bits & EXTENT_DELALLOC)) WARN_ON(1); /* * set_bit and clear bit hooks normally require _irqsave/restore * but in this case, we are only testing for the DELALLOC * bit, which is only set or cleared with irqs on */ - if (!(state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { + if (!(state->state & EXTENT_DELALLOC) && (bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; - u32 num_extents = count_max_extents(len); + u32 num_extents = count_max_extents(fs_info, len); bool do_list = !btrfs_is_free_space_inode(BTRFS_I(inode)); spin_lock(&BTRFS_I(inode)->lock); @@ -1933,7 +2417,7 @@ void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state, fs_info->delalloc_batch); spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->delalloc_bytes += len; - if (*bits & EXTENT_DEFRAG) + if (bits & EXTENT_DEFRAG) BTRFS_I(inode)->defrag_bytes += len; if (do_list && !test_bit(BTRFS_INODE_IN_DELALLOC_LIST, &BTRFS_I(inode)->runtime_flags)) @@ -1942,7 +2426,7 @@ void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state, } if (!(state->state & EXTENT_DELALLOC_NEW) && - (*bits & EXTENT_DELALLOC_NEW)) { + (bits & EXTENT_DELALLOC_NEW)) { spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->new_delalloc_bytes += state->end + 1 - state->start; @@ -1955,14 +2439,14 @@ void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state, * accounting happens. */ void btrfs_clear_delalloc_extent(struct inode *vfs_inode, - struct extent_state *state, unsigned *bits) + struct extent_state *state, u32 bits) { struct btrfs_inode *inode = BTRFS_I(vfs_inode); struct btrfs_fs_info *fs_info = btrfs_sb(vfs_inode->i_sb); u64 len = state->end + 1 - state->start; - u32 num_extents = count_max_extents(len); + u32 num_extents = count_max_extents(fs_info, len); - if ((state->state & EXTENT_DEFRAG) && (*bits & EXTENT_DEFRAG)) { + if ((state->state & EXTENT_DEFRAG) && (bits & EXTENT_DEFRAG)) { spin_lock(&inode->lock); inode->defrag_bytes -= len; spin_unlock(&inode->lock); @@ -1973,7 +2457,7 @@ void btrfs_clear_delalloc_extent(struct inode *vfs_inode, * but in this case, we are only testing for the DELALLOC * bit, which is only set or cleared with irqs on */ - if ((state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { + if ((state->state & EXTENT_DELALLOC) && (bits & EXTENT_DELALLOC)) { struct btrfs_root *root = inode->root; bool do_list = !btrfs_is_free_space_inode(inode); @@ -1986,7 +2470,7 @@ void btrfs_clear_delalloc_extent(struct inode *vfs_inode, * don't need to call delalloc_release_metadata if there is an * error. */ - if (*bits & EXTENT_CLEAR_META_RESV && + if (bits & EXTENT_CLEAR_META_RESV && root != fs_info->tree_root) btrfs_delalloc_release_metadata(inode, len, false); @@ -1994,12 +2478,10 @@ void btrfs_clear_delalloc_extent(struct inode *vfs_inode, if (btrfs_is_testing(fs_info)) return; - if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID && + if (!btrfs_is_data_reloc_root(root) && do_list && !(state->state & EXTENT_NORESERVE) && - (*bits & EXTENT_CLEAR_DATA_RESV)) - btrfs_free_reserved_data_space_noquota( - &inode->vfs_inode, - state->start, len); + (bits & EXTENT_CLEAR_DATA_RESV)) + btrfs_free_reserved_data_space_noquota(fs_info, len); percpu_counter_add_batch(&fs_info->delalloc_bytes, -len, fs_info->delalloc_batch); @@ -2013,55 +2495,17 @@ void btrfs_clear_delalloc_extent(struct inode *vfs_inode, } if ((state->state & EXTENT_DELALLOC_NEW) && - (*bits & EXTENT_DELALLOC_NEW)) { + (bits & EXTENT_DELALLOC_NEW)) { spin_lock(&inode->lock); ASSERT(inode->new_delalloc_bytes >= len); inode->new_delalloc_bytes -= len; + if (bits & EXTENT_ADD_INODE_BYTES) + inode_add_bytes(&inode->vfs_inode, len); spin_unlock(&inode->lock); } } /* - * btrfs_bio_fits_in_stripe - Checks whether the size of the given bio will fit - * in a chunk's stripe. This function ensures that bios do not span a - * stripe/chunk - * - * @page - The page we are about to add to the bio - * @size - size we want to add to the bio - * @bio - bio we want to ensure is smaller than a stripe - * @bio_flags - flags of the bio - * - * return 1 if page cannot be added to the bio - * return 0 if page can be added to the bio - * return error otherwise - */ -int btrfs_bio_fits_in_stripe(struct page *page, size_t size, struct bio *bio, - unsigned long bio_flags) -{ - 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; - int ret; - struct btrfs_io_geometry geom; - - if (bio_flags & EXTENT_BIO_COMPRESSED) - return 0; - - length = bio->bi_iter.bi_size; - map_length = length; - ret = btrfs_get_io_geometry(fs_info, btrfs_op(bio), logical, map_length, - &geom); - if (ret < 0) - return ret; - - if (geom.len < length + size) - return 1; - return 0; -} - -/* * in order to insert checksums into the metadata in large chunks, * we wait until bio submission time. All the pages in the bio are * checksummed and sums are attached onto the ordered extent record. @@ -2069,107 +2513,275 @@ int btrfs_bio_fits_in_stripe(struct page *page, size_t size, struct bio *bio, * At IO completion time the cums attached on the ordered extent record * are inserted into the btree */ -static blk_status_t btrfs_submit_bio_start(void *private_data, struct bio *bio, - u64 bio_offset) +static blk_status_t btrfs_submit_bio_start(struct inode *inode, struct bio *bio, + u64 dio_file_offset) { - struct inode *inode = private_data; - blk_status_t ret = 0; - - ret = btrfs_csum_one_bio(inode, bio, 0, 0); - BUG_ON(ret); /* -ENOMEM */ - return 0; + return btrfs_csum_one_bio(BTRFS_I(inode), bio, (u64)-1, false); } /* - * extent_io.c submission hook. This does the right thing for csum calculation - * on write, or reading the csums from the tree before a read. - * - * Rules about async/sync submit, - * a) read: sync submit - * - * b) write without checksum: sync submit + * Split an extent_map at [start, start + len] * - * c) write with checksum: - * c-1) if bio is issued by fsync: sync submit - * (sync_writers != 0) - * - * c-2) if root is reloc root: sync submit - * (only in case of buffered IO) - * - * c-3) otherwise: async submit + * This function is intended to be used only for extract_ordered_extent(). */ -static blk_status_t btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, - int mirror_num, - unsigned long bio_flags) +static int split_zoned_em(struct btrfs_inode *inode, u64 start, u64 len, + u64 pre, u64 post) +{ + struct extent_map_tree *em_tree = &inode->extent_tree; + struct extent_map *em; + struct extent_map *split_pre = NULL; + struct extent_map *split_mid = NULL; + struct extent_map *split_post = NULL; + int ret = 0; + unsigned long flags; + + /* Sanity check */ + if (pre == 0 && post == 0) + return 0; + + split_pre = alloc_extent_map(); + if (pre) + split_mid = alloc_extent_map(); + if (post) + split_post = alloc_extent_map(); + if (!split_pre || (pre && !split_mid) || (post && !split_post)) { + ret = -ENOMEM; + goto out; + } + + ASSERT(pre + post < len); + + lock_extent(&inode->io_tree, start, start + len - 1, NULL); + write_lock(&em_tree->lock); + em = lookup_extent_mapping(em_tree, start, len); + if (!em) { + ret = -EIO; + goto out_unlock; + } + + ASSERT(em->len == len); + ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)); + ASSERT(em->block_start < EXTENT_MAP_LAST_BYTE); + ASSERT(test_bit(EXTENT_FLAG_PINNED, &em->flags)); + ASSERT(!test_bit(EXTENT_FLAG_LOGGING, &em->flags)); + ASSERT(!list_empty(&em->list)); + + flags = em->flags; + clear_bit(EXTENT_FLAG_PINNED, &em->flags); + + /* First, replace the em with a new extent_map starting from * em->start */ + split_pre->start = em->start; + split_pre->len = (pre ? pre : em->len - post); + split_pre->orig_start = split_pre->start; + split_pre->block_start = em->block_start; + split_pre->block_len = split_pre->len; + split_pre->orig_block_len = split_pre->block_len; + split_pre->ram_bytes = split_pre->len; + split_pre->flags = flags; + split_pre->compress_type = em->compress_type; + split_pre->generation = em->generation; + + replace_extent_mapping(em_tree, em, split_pre, 1); + + /* + * Now we only have an extent_map at: + * [em->start, em->start + pre] if pre != 0 + * [em->start, em->start + em->len - post] if pre == 0 + */ + + if (pre) { + /* Insert the middle extent_map */ + split_mid->start = em->start + pre; + split_mid->len = em->len - pre - post; + split_mid->orig_start = split_mid->start; + split_mid->block_start = em->block_start + pre; + split_mid->block_len = split_mid->len; + split_mid->orig_block_len = split_mid->block_len; + split_mid->ram_bytes = split_mid->len; + split_mid->flags = flags; + split_mid->compress_type = em->compress_type; + split_mid->generation = em->generation; + add_extent_mapping(em_tree, split_mid, 1); + } + + if (post) { + split_post->start = em->start + em->len - post; + split_post->len = post; + split_post->orig_start = split_post->start; + split_post->block_start = em->block_start + em->len - post; + split_post->block_len = split_post->len; + split_post->orig_block_len = split_post->block_len; + split_post->ram_bytes = split_post->len; + split_post->flags = flags; + split_post->compress_type = em->compress_type; + split_post->generation = em->generation; + add_extent_mapping(em_tree, split_post, 1); + } + + /* Once for us */ + free_extent_map(em); + /* Once for the tree */ + free_extent_map(em); + +out_unlock: + write_unlock(&em_tree->lock); + unlock_extent(&inode->io_tree, start, start + len - 1, NULL); +out: + free_extent_map(split_pre); + free_extent_map(split_mid); + free_extent_map(split_post); + + return ret; +} +static blk_status_t extract_ordered_extent(struct btrfs_inode *inode, + struct bio *bio, loff_t file_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; - blk_status_t ret = 0; - int skip_sum; - int async = !atomic_read(&BTRFS_I(inode)->sync_writers); + struct btrfs_ordered_extent *ordered; + u64 start = (u64)bio->bi_iter.bi_sector << SECTOR_SHIFT; + u64 file_len; + u64 len = bio->bi_iter.bi_size; + u64 end = start + len; + u64 ordered_end; + u64 pre, post; + int ret = 0; - skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; + ordered = btrfs_lookup_ordered_extent(inode, file_offset); + if (WARN_ON_ONCE(!ordered)) + return BLK_STS_IOERR; - if (btrfs_is_free_space_inode(BTRFS_I(inode))) - metadata = BTRFS_WQ_ENDIO_FREE_SPACE; + /* No need to split */ + if (ordered->disk_num_bytes == len) + goto out; - if (bio_op(bio) != REQ_OP_WRITE) { - ret = btrfs_bio_wq_end_io(fs_info, bio, metadata); - if (ret) - goto out; + /* We cannot split once end_bio'd ordered extent */ + if (WARN_ON_ONCE(ordered->bytes_left != ordered->disk_num_bytes)) { + ret = -EINVAL; + goto out; + } - if (bio_flags & EXTENT_BIO_COMPRESSED) { - ret = btrfs_submit_compressed_read(inode, bio, - mirror_num, - bio_flags); - goto out; - } else if (!skip_sum) { - ret = btrfs_lookup_bio_sums(inode, bio, (u64)-1, NULL); - if (ret) - goto out; - } - goto mapit; - } else if (async && !skip_sum) { - /* csum items have already been cloned */ - if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID) - goto mapit; - /* we're doing a write, do the async checksumming */ - ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, bio_flags, - 0, inode, btrfs_submit_bio_start); + /* We cannot split a compressed ordered extent */ + if (WARN_ON_ONCE(ordered->disk_num_bytes != ordered->num_bytes)) { + ret = -EINVAL; + goto out; + } + + ordered_end = ordered->disk_bytenr + ordered->disk_num_bytes; + /* bio must be in one ordered extent */ + if (WARN_ON_ONCE(start < ordered->disk_bytenr || end > ordered_end)) { + ret = -EINVAL; + goto out; + } + + /* Checksum list should be empty */ + if (WARN_ON_ONCE(!list_empty(&ordered->list))) { + ret = -EINVAL; goto out; - } else if (!skip_sum) { - ret = btrfs_csum_one_bio(inode, bio, 0, 0); - if (ret) - goto out; } -mapit: - ret = btrfs_map_bio(fs_info, bio, mirror_num); + file_len = ordered->num_bytes; + pre = start - ordered->disk_bytenr; + post = ordered_end - end; + + ret = btrfs_split_ordered_extent(ordered, pre, post); + if (ret) + goto out; + ret = split_zoned_em(inode, file_offset, file_len, pre, post); out: + btrfs_put_ordered_extent(ordered); + + return errno_to_blk_status(ret); +} + +void btrfs_submit_data_write_bio(struct inode *inode, struct bio *bio, int mirror_num) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct btrfs_inode *bi = BTRFS_I(inode); + blk_status_t ret; + + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + ret = extract_ordered_extent(bi, bio, + page_offset(bio_first_bvec_all(bio)->bv_page)); + if (ret) { + btrfs_bio_end_io(btrfs_bio(bio), ret); + return; + } + } + + /* + * If we need to checksum, and the I/O is not issued by fsync and + * friends, that is ->sync_writers != 0, defer the submission to a + * workqueue to parallelize it. + * + * Csum items for reloc roots have already been cloned at this point, + * so they are handled as part of the no-checksum case. + */ + if (!(bi->flags & BTRFS_INODE_NODATASUM) && + !test_bit(BTRFS_FS_STATE_NO_CSUMS, &fs_info->fs_state) && + !btrfs_is_data_reloc_root(bi->root)) { + if (!atomic_read(&bi->sync_writers) && + btrfs_wq_submit_bio(inode, bio, mirror_num, 0, + btrfs_submit_bio_start)) + return; + + ret = btrfs_csum_one_bio(bi, bio, (u64)-1, false); + if (ret) { + btrfs_bio_end_io(btrfs_bio(bio), ret); + return; + } + } + btrfs_submit_bio(fs_info, bio, mirror_num); +} + +void btrfs_submit_data_read_bio(struct inode *inode, struct bio *bio, + int mirror_num, enum btrfs_compression_type compress_type) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + blk_status_t ret; + + if (compress_type != BTRFS_COMPRESS_NONE) { + /* + * btrfs_submit_compressed_read will handle completing the bio + * if there were any errors, so just return here. + */ + btrfs_submit_compressed_read(inode, bio, mirror_num); + return; + } + + /* Save the original iter for read repair */ + btrfs_bio(bio)->iter = bio->bi_iter; + + /* + * Lookup bio sums does extra checks around whether we need to csum or + * not, which is why we ignore skip_sum here. + */ + ret = btrfs_lookup_bio_sums(inode, bio, NULL); if (ret) { - bio->bi_status = ret; - bio_endio(bio); + btrfs_bio_end_io(btrfs_bio(bio), ret); + return; } - return ret; + + btrfs_submit_bio(fs_info, bio, mirror_num); } /* * given a list of ordered sums record them in the inode. This happens * 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, struct list_head *list) +static int add_pending_csums(struct btrfs_trans_handle *trans, + struct list_head *list) { struct btrfs_ordered_sum *sum; + struct btrfs_root *csum_root = NULL; int ret; list_for_each_entry(sum, list, list) { trans->adding_csums = true; - ret = btrfs_csum_file_blocks(trans, - BTRFS_I(inode)->root->fs_info->csum_root, sum); + if (!csum_root) + csum_root = btrfs_csum_root(trans->fs_info, + sum->bytenr); + ret = btrfs_csum_file_blocks(trans, csum_root, sum); trans->adding_csums = false; if (ret) return ret; @@ -2177,13 +2789,71 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans, return 0; } -int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end, +static int btrfs_find_new_delalloc_bytes(struct btrfs_inode *inode, + const u64 start, + const u64 len, + struct extent_state **cached_state) +{ + u64 search_start = start; + const u64 end = start + len - 1; + + while (search_start < end) { + const u64 search_len = end - search_start + 1; + struct extent_map *em; + u64 em_len; + int ret = 0; + + em = btrfs_get_extent(inode, NULL, 0, search_start, search_len); + if (IS_ERR(em)) + return PTR_ERR(em); + + if (em->block_start != EXTENT_MAP_HOLE) + goto next; + + em_len = em->len; + if (em->start < search_start) + em_len -= search_start - em->start; + if (em_len > search_len) + em_len = search_len; + + ret = set_extent_bit(&inode->io_tree, search_start, + search_start + em_len - 1, + EXTENT_DELALLOC_NEW, cached_state, + GFP_NOFS); +next: + search_start = extent_map_end(em); + free_extent_map(em); + if (ret) + return ret; + } + return 0; +} + +int btrfs_set_extent_delalloc(struct btrfs_inode *inode, u64 start, u64 end, unsigned int extra_bits, struct extent_state **cached_state) { WARN_ON(PAGE_ALIGNED(end)); - return set_extent_delalloc(&BTRFS_I(inode)->io_tree, start, end, - extra_bits, cached_state); + + if (start >= i_size_read(&inode->vfs_inode) && + !(inode->flags & BTRFS_INODE_PREALLOC)) { + /* + * There can't be any extents following eof in this case so just + * set the delalloc new bit for the range directly. + */ + extra_bits |= EXTENT_DELALLOC_NEW; + } else { + int ret; + + ret = btrfs_find_new_delalloc_bytes(inode, start, + end + 1 - start, + cached_state); + if (ret) + return ret; + } + + return set_extent_delalloc(&inode->io_tree, start, end, extra_bits, + cached_state); } /* see btrfs_writepage_start_hook for details on why this is required */ @@ -2200,7 +2870,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work) struct extent_state *cached_state = NULL; struct extent_changeset *data_reserved = NULL; struct page *page; - struct inode *inode; + struct btrfs_inode *inode; u64 page_start; u64 page_end; int ret = 0; @@ -2208,7 +2878,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work) fixup = container_of(work, struct btrfs_writepage_fixup, work); page = fixup->page; - inode = fixup->inode; + inode = BTRFS_I(fixup->inode); page_start = page_offset(page); page_end = page_offset(page) + PAGE_SIZE - 1; @@ -2245,8 +2915,7 @@ again: * when the page was already properly dealt with. */ if (!ret) { - btrfs_delalloc_release_extents(BTRFS_I(inode), - PAGE_SIZE); + btrfs_delalloc_release_extents(inode, PAGE_SIZE); btrfs_delalloc_release_space(inode, data_reserved, page_start, PAGE_SIZE, true); @@ -2262,20 +2931,18 @@ again: if (ret) goto out_page; - lock_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end, - &cached_state); + lock_extent(&inode->io_tree, page_start, page_end, &cached_state); /* already ordered? We're done */ - if (PagePrivate2(page)) + if (PageOrdered(page)) goto out_reserved; - ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start, - PAGE_SIZE); + ordered = btrfs_lookup_ordered_range(inode, page_start, PAGE_SIZE); if (ordered) { - unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, - page_end, &cached_state); + unlock_extent(&inode->io_tree, page_start, page_end, + &cached_state); unlock_page(page); - btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_start_ordered_extent(ordered, 1); btrfs_put_ordered_extent(ordered); goto again; } @@ -2295,12 +2962,11 @@ again: BUG_ON(!PageDirty(page)); free_delalloc_space = false; out_reserved: - btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); + btrfs_delalloc_release_extents(inode, PAGE_SIZE); if (free_delalloc_space) btrfs_delalloc_release_space(inode, data_reserved, page_start, PAGE_SIZE, true); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end, - &cached_state); + unlock_extent(&inode->io_tree, page_start, page_end, &cached_state); out_page: if (ret) { /* @@ -2312,7 +2978,7 @@ out_page: clear_page_dirty_for_io(page); SetPageError(page); } - ClearPageChecked(page); + btrfs_page_clear_checked(inode->root->fs_info, page, page_start, PAGE_SIZE); unlock_page(page); put_page(page); kfree(fixup); @@ -2322,7 +2988,7 @@ out_page: * that could need flushing space. Recursing back to fixup worker would * deadlock. */ - btrfs_add_delayed_iput(inode); + btrfs_add_delayed_iput(&inode->vfs_inode); } /* @@ -2336,14 +3002,14 @@ out_page: * to fix it up. The async helper will wait for ordered extents, set * the delalloc bit and make it safe to write the page. */ -int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end) +int btrfs_writepage_cow_fixup(struct page *page) { struct inode *inode = page->mapping->host; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_writepage_fixup *fixup; - /* this page is properly in the ordered list */ - if (TestClearPagePrivate2(page)) + /* This page has ordered extent covering it already */ + if (PageOrdered(page)) return 0; /* @@ -2367,7 +3033,7 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end) * page->mapping outside of the page lock. */ ihold(inode); - SetPageChecked(page); + btrfs_page_set_checked(fs_info, page, page_offset(page), PAGE_SIZE); get_page(page); btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL); fixup->page = page; @@ -2378,19 +3044,22 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end) } static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, - struct inode *inode, u64 file_pos, - u64 disk_bytenr, u64 disk_num_bytes, - u64 num_bytes, u64 ram_bytes, - u8 compression, u8 encryption, - u16 other_encoding, int extent_type) + struct btrfs_inode *inode, u64 file_pos, + struct btrfs_file_extent_item *stack_fi, + const bool update_inode_bytes, + u64 qgroup_reserved) { - struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_file_extent_item *fi; + struct btrfs_root *root = inode->root; + const u64 sectorsize = root->fs_info->sectorsize; struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_key ins; - u64 qg_released; - int extent_inserted = 0; + u64 disk_num_bytes = btrfs_stack_file_extent_disk_num_bytes(stack_fi); + u64 disk_bytenr = btrfs_stack_file_extent_disk_bytenr(stack_fi); + u64 offset = btrfs_stack_file_extent_offset(stack_fi); + u64 num_bytes = btrfs_stack_file_extent_num_bytes(stack_fi); + u64 ram_bytes = btrfs_stack_file_extent_ram_bytes(stack_fi); + struct btrfs_drop_extents_args drop_args = { 0 }; int ret; path = btrfs_alloc_path(); @@ -2406,57 +3075,64 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, * the caller is expected to unpin it and allow it to be merged * with the others. */ - ret = __btrfs_drop_extents(trans, root, inode, path, file_pos, - file_pos + num_bytes, NULL, 0, - 1, sizeof(*fi), &extent_inserted); + drop_args.path = path; + drop_args.start = file_pos; + drop_args.end = file_pos + num_bytes; + drop_args.replace_extent = true; + drop_args.extent_item_size = sizeof(*stack_fi); + ret = btrfs_drop_extents(trans, root, inode, &drop_args); if (ret) goto out; - if (!extent_inserted) { - ins.objectid = btrfs_ino(BTRFS_I(inode)); + if (!drop_args.extent_inserted) { + ins.objectid = btrfs_ino(inode); ins.offset = file_pos; ins.type = BTRFS_EXTENT_DATA_KEY; - path->leave_spinning = 1; ret = btrfs_insert_empty_item(trans, root, path, &ins, - sizeof(*fi)); + sizeof(*stack_fi)); if (ret) goto out; } leaf = path->nodes[0]; - fi = btrfs_item_ptr(leaf, path->slots[0], - struct btrfs_file_extent_item); - btrfs_set_file_extent_generation(leaf, fi, trans->transid); - btrfs_set_file_extent_type(leaf, fi, extent_type); - btrfs_set_file_extent_disk_bytenr(leaf, fi, disk_bytenr); - btrfs_set_file_extent_disk_num_bytes(leaf, fi, disk_num_bytes); - btrfs_set_file_extent_offset(leaf, fi, 0); - btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes); - btrfs_set_file_extent_ram_bytes(leaf, fi, ram_bytes); - btrfs_set_file_extent_compression(leaf, fi, compression); - btrfs_set_file_extent_encryption(leaf, fi, encryption); - btrfs_set_file_extent_other_encoding(leaf, fi, other_encoding); + btrfs_set_stack_file_extent_generation(stack_fi, trans->transid); + write_extent_buffer(leaf, stack_fi, + btrfs_item_ptr_offset(leaf, path->slots[0]), + sizeof(struct btrfs_file_extent_item)); btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); - inode_add_bytes(inode, num_bytes); + /* + * If we dropped an inline extent here, we know the range where it is + * was not marked with the EXTENT_DELALLOC_NEW bit, so we update the + * number of bytes only for that range containing the inline extent. + * The remaining of the range will be processed when clearning the + * EXTENT_DELALLOC_BIT bit through the ordered extent completion. + */ + if (file_pos == 0 && !IS_ALIGNED(drop_args.bytes_found, sectorsize)) { + u64 inline_size = round_down(drop_args.bytes_found, sectorsize); + + inline_size = drop_args.bytes_found - inline_size; + btrfs_update_inode_bytes(inode, sectorsize, inline_size); + drop_args.bytes_found -= inline_size; + num_bytes -= sectorsize; + } + + if (update_inode_bytes) + btrfs_update_inode_bytes(inode, num_bytes, drop_args.bytes_found); ins.objectid = disk_bytenr; ins.offset = disk_num_bytes; ins.type = BTRFS_EXTENT_ITEM_KEY; - /* - * Release the reserved range from inode dirty range map, as it is - * already moved into delayed_ref_head - */ - ret = btrfs_qgroup_release_data(inode, file_pos, ram_bytes); - if (ret < 0) + ret = btrfs_inode_set_file_extent_range(inode, file_pos, ram_bytes); + if (ret) goto out; - qg_released = ret; - ret = btrfs_alloc_reserved_file_extent(trans, root, - btrfs_ino(BTRFS_I(inode)), - file_pos, qg_released, &ins); + + ret = btrfs_alloc_reserved_file_extent(trans, root, btrfs_ino(inode), + file_pos - offset, + qgroup_reserved, &ins); out: btrfs_free_path(path); @@ -2478,17 +3154,56 @@ static void btrfs_release_delalloc_bytes(struct btrfs_fs_info *fs_info, btrfs_put_block_group(cache); } -/* as ordered data IO finishes, this gets called so we can finish +static int insert_ordered_extent_file_extent(struct btrfs_trans_handle *trans, + struct btrfs_ordered_extent *oe) +{ + struct btrfs_file_extent_item stack_fi; + bool update_inode_bytes; + u64 num_bytes = oe->num_bytes; + u64 ram_bytes = oe->ram_bytes; + + memset(&stack_fi, 0, sizeof(stack_fi)); + btrfs_set_stack_file_extent_type(&stack_fi, BTRFS_FILE_EXTENT_REG); + btrfs_set_stack_file_extent_disk_bytenr(&stack_fi, oe->disk_bytenr); + btrfs_set_stack_file_extent_disk_num_bytes(&stack_fi, + oe->disk_num_bytes); + btrfs_set_stack_file_extent_offset(&stack_fi, oe->offset); + if (test_bit(BTRFS_ORDERED_TRUNCATED, &oe->flags)) { + num_bytes = oe->truncated_len; + ram_bytes = num_bytes; + } + btrfs_set_stack_file_extent_num_bytes(&stack_fi, num_bytes); + btrfs_set_stack_file_extent_ram_bytes(&stack_fi, ram_bytes); + btrfs_set_stack_file_extent_compression(&stack_fi, oe->compress_type); + /* Encryption and other encoding is reserved and all 0 */ + + /* + * For delalloc, when completing an ordered extent we update the inode's + * bytes when clearing the range in the inode's io tree, so pass false + * as the argument 'update_inode_bytes' to insert_reserved_file_extent(), + * except if the ordered extent was truncated. + */ + update_inode_bytes = test_bit(BTRFS_ORDERED_DIRECT, &oe->flags) || + test_bit(BTRFS_ORDERED_ENCODED, &oe->flags) || + test_bit(BTRFS_ORDERED_TRUNCATED, &oe->flags); + + return insert_reserved_file_extent(trans, BTRFS_I(oe->inode), + oe->file_offset, &stack_fi, + update_inode_bytes, oe->qgroup_rsv); +} + +/* + * As ordered data IO finishes, this gets called so we can finish * an ordered extent if the range of bytes in the file it covers are * fully written. */ -static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) +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_inode *inode = BTRFS_I(ordered_extent->inode); + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_trans_handle *trans = NULL; - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; + struct extent_io_tree *io_tree = &inode->io_tree; struct extent_state *cached_state = NULL; u64 start, end; int compress_type = 0; @@ -2496,27 +3211,35 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) u64 logical_len = ordered_extent->num_bytes; bool freespace_inode; bool truncated = false; - bool range_locked = false; - bool clear_new_delalloc_bytes = false; bool clear_reserved_extent = true; - unsigned int clear_bits; + unsigned int clear_bits = EXTENT_DEFRAG; start = ordered_extent->file_offset; end = start + ordered_extent->num_bytes - 1; if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) && !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags) && - !test_bit(BTRFS_ORDERED_DIRECT, &ordered_extent->flags)) - clear_new_delalloc_bytes = true; + !test_bit(BTRFS_ORDERED_DIRECT, &ordered_extent->flags) && + !test_bit(BTRFS_ORDERED_ENCODED, &ordered_extent->flags)) + clear_bits |= EXTENT_DELALLOC_NEW; - freespace_inode = btrfs_is_free_space_inode(BTRFS_I(inode)); + freespace_inode = btrfs_is_free_space_inode(inode); + if (!freespace_inode) + btrfs_lockdep_acquire(fs_info, btrfs_ordered_extent); if (test_bit(BTRFS_ORDERED_IOERR, &ordered_extent->flags)) { ret = -EIO; goto out; } - btrfs_free_io_failure_record(BTRFS_I(inode), start, end); + /* A valid bdev implies a write on a sequential zone */ + if (ordered_extent->bdev) { + btrfs_rewrite_logical_zoned(ordered_extent); + btrfs_zone_finish_endio(fs_info, ordered_extent->disk_bytenr, + ordered_extent->disk_num_bytes); + } + + btrfs_free_io_failure_record(inode, start, end); if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) { truncated = true; @@ -2529,14 +3252,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) { BUG_ON(!list_empty(&ordered_extent->list)); /* Logic error */ - /* - * For mwrite(mmap + memset to write) case, we still reserve - * space for NOCOW range. - * As NOCOW won't cause a new delayed ref, just free the space - */ - btrfs_qgroup_free_data(inode, NULL, start, - ordered_extent->num_bytes); - btrfs_ordered_update_i_size(inode, 0, ordered_extent); + btrfs_inode_safe_disk_i_size_write(inode, 0); if (freespace_inode) trans = btrfs_join_transaction_spacecache(root); else @@ -2546,15 +3262,15 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) trans = NULL; goto out; } - trans->block_rsv = &BTRFS_I(inode)->block_rsv; + trans->block_rsv = &inode->block_rsv; ret = btrfs_update_inode_fallback(trans, root, inode); if (ret) /* -ENOMEM or corruption */ btrfs_abort_transaction(trans, ret); goto out; } - range_locked = true; - lock_extent_bits(io_tree, start, end, &cached_state); + clear_bits |= EXTENT_LOCKED; + lock_extent(io_tree, start, end, &cached_state); if (freespace_inode) trans = btrfs_join_transaction_spacecache(root); @@ -2566,26 +3282,21 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out; } - trans->block_rsv = &BTRFS_I(inode)->block_rsv; + trans->block_rsv = &inode->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); - btrfs_qgroup_free_data(inode, NULL, start, - ordered_extent->num_bytes); - ret = btrfs_mark_extent_written(trans, BTRFS_I(inode), + ret = btrfs_mark_extent_written(trans, inode, ordered_extent->file_offset, ordered_extent->file_offset + logical_len); + btrfs_zoned_release_data_reloc_bg(fs_info, ordered_extent->disk_bytenr, + ordered_extent->disk_num_bytes); } else { BUG_ON(root == fs_info->tree_root); - ret = insert_reserved_file_extent(trans, inode, start, - ordered_extent->disk_bytenr, - ordered_extent->disk_num_bytes, - logical_len, logical_len, - compress_type, 0, 0, - BTRFS_FILE_EXTENT_REG); + ret = insert_ordered_extent_file_extent(trans, ordered_extent); if (!ret) { clear_reserved_extent = false; btrfs_release_delalloc_bytes(fs_info, @@ -2593,21 +3304,31 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) ordered_extent->disk_num_bytes); } } - unpin_extent_cache(&BTRFS_I(inode)->extent_tree, - ordered_extent->file_offset, + unpin_extent_cache(&inode->extent_tree, ordered_extent->file_offset, ordered_extent->num_bytes, trans->transid); if (ret < 0) { btrfs_abort_transaction(trans, ret); goto out; } - ret = add_pending_csums(trans, inode, &ordered_extent->list); + ret = add_pending_csums(trans, &ordered_extent->list); if (ret) { btrfs_abort_transaction(trans, ret); goto out; } - btrfs_ordered_update_i_size(inode, 0, ordered_extent); + /* + * If this is a new delalloc range, clear its new delalloc flag to + * update the inode's number of bytes. This needs to be done first + * before updating the inode item. + */ + if ((clear_bits & EXTENT_DELALLOC_NEW) && + !test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) + clear_extent_bit(&inode->io_tree, start, end, + EXTENT_DELALLOC_NEW | EXTENT_ADD_INODE_BYTES, + &cached_state); + + btrfs_inode_safe_disk_i_size_write(inode, 0); ret = btrfs_update_inode_fallback(trans, root, inode); if (ret) { /* -ENOMEM or corruption */ btrfs_abort_transaction(trans, ret); @@ -2615,13 +3336,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) } ret = 0; out: - clear_bits = EXTENT_DEFRAG; - if (range_locked) - clear_bits |= EXTENT_LOCKED; - if (clear_new_delalloc_bytes) - clear_bits |= EXTENT_DELALLOC_NEW; - clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits, - (clear_bits & EXTENT_LOCKED) ? 1 : 0, 0, + clear_extent_bit(&inode->io_tree, start, end, clear_bits, &cached_state); if (trans) @@ -2630,12 +3345,24 @@ out: if (ret || truncated) { u64 unwritten_start = start; + /* + * If we failed to finish this ordered extent for any reason we + * need to make sure BTRFS_ORDERED_IOERR is set on the ordered + * extent, and mark the inode with the error if it wasn't + * already set. Any error during writeback would have already + * set the mapping error, so we need to set it if we're the ones + * marking this ordered extent as failed. + */ + if (ret && !test_and_set_bit(BTRFS_ORDERED_IOERR, + &ordered_extent->flags)) + mapping_set_error(ordered_extent->inode->i_mapping, -EIO); + if (truncated) unwritten_start += logical_len; clear_extent_uptodate(io_tree, unwritten_start, end, NULL); - /* Drop the cache for the part of the extent we didn't write. */ - btrfs_drop_extent_cache(BTRFS_I(inode), unwritten_start, end, 0); + /* Drop extent maps for the part of the extent we didn't write. */ + btrfs_drop_extent_map_range(inode, unwritten_start, end, false); /* * If the ordered extent had an IOERR or something else went @@ -2680,103 +3407,150 @@ out: return ret; } -static void finish_ordered_fn(struct btrfs_work *work) +void btrfs_writepage_endio_finish_ordered(struct btrfs_inode *inode, + struct page *page, u64 start, + u64 end, bool uptodate) { - struct btrfs_ordered_extent *ordered_extent; - ordered_extent = container_of(work, struct btrfs_ordered_extent, work); - btrfs_finish_ordered_io(ordered_extent); + trace_btrfs_writepage_end_io_hook(inode, start, end, uptodate); + + btrfs_mark_ordered_io_finished(inode, page, start, end + 1 - start, uptodate); } -void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start, - u64 end, int uptodate) +/* + * Verify the checksum for a single sector without any extra action that depend + * on the type of I/O. + */ +int btrfs_check_sector_csum(struct btrfs_fs_info *fs_info, struct page *page, + u32 pgoff, u8 *csum, const u8 * const csum_expected) { - struct inode *inode = page->mapping->host; - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_ordered_extent *ordered_extent = NULL; - struct btrfs_workqueue *wq; + SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); + char *kaddr; - trace_btrfs_writepage_end_io_hook(page, start, end, uptodate); + ASSERT(pgoff + fs_info->sectorsize <= PAGE_SIZE); - ClearPagePrivate2(page); - if (!btrfs_dec_test_ordered_pending(inode, &ordered_extent, start, - end - start + 1, uptodate)) - return; + shash->tfm = fs_info->csum_shash; - if (btrfs_is_free_space_inode(BTRFS_I(inode))) - wq = fs_info->endio_freespace_worker; - else - wq = fs_info->endio_write_workers; + kaddr = kmap_local_page(page) + pgoff; + crypto_shash_digest(shash, kaddr, fs_info->sectorsize, csum); + kunmap_local(kaddr); + + if (memcmp(csum, csum_expected, fs_info->csum_size)) + return -EIO; + return 0; +} + +static u8 *btrfs_csum_ptr(const struct btrfs_fs_info *fs_info, u8 *csums, u64 offset) +{ + u64 offset_in_sectors = offset >> fs_info->sectorsize_bits; - btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL); - btrfs_queue_work(wq, &ordered_extent->work); + return csums + offset_in_sectors * fs_info->csum_size; } -static int __readpage_endio_check(struct inode *inode, - struct btrfs_io_bio *io_bio, - int icsum, struct page *page, - int pgoff, u64 start, size_t len) +/* + * check_data_csum - verify checksum of one sector of uncompressed data + * @inode: inode + * @bbio: btrfs_bio which contains the csum + * @bio_offset: offset to the beginning of the bio (in bytes) + * @page: page where is the data to be verified + * @pgoff: offset inside the page + * + * The length of such check is always one sector size. + * + * When csum mismatch is detected, we will also report the error and fill the + * corrupted range with zero. (Thus it needs the extra parameters) + */ +int btrfs_check_data_csum(struct inode *inode, struct btrfs_bio *bbio, + u32 bio_offset, struct page *page, u32 pgoff) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); - char *kaddr; - u16 csum_size = btrfs_super_csum_size(fs_info->super_copy); + u32 len = fs_info->sectorsize; u8 *csum_expected; u8 csum[BTRFS_CSUM_SIZE]; - csum_expected = ((u8 *)io_bio->csum) + icsum * csum_size; + ASSERT(pgoff + len <= PAGE_SIZE); - kaddr = kmap_atomic(page); - shash->tfm = fs_info->csum_shash; - - crypto_shash_init(shash); - crypto_shash_update(shash, kaddr + pgoff, len); - crypto_shash_final(shash, csum); + csum_expected = btrfs_csum_ptr(fs_info, bbio->csum, bio_offset); - if (memcmp(csum, csum_expected, csum_size)) + if (btrfs_check_sector_csum(fs_info, page, pgoff, csum, csum_expected)) goto zeroit; - - kunmap_atomic(kaddr); return 0; + zeroit: - 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); + btrfs_print_data_csum_error(BTRFS_I(inode), + bbio->file_offset + bio_offset, + csum, csum_expected, bbio->mirror_num); + if (bbio->device) + btrfs_dev_stat_inc_and_print(bbio->device, + BTRFS_DEV_STAT_CORRUPTION_ERRS); + memzero_page(page, pgoff, len); return -EIO; } /* - * when reads are done, we need to check csums to verify the data is correct + * When reads are done, we need to check csums to verify the data is correct. * if there's a match, we allow the bio to finish. If not, the code in * extent_io.c will try to find good copies for us. + * + * @bio_offset: offset to the beginning of the bio (in bytes) + * @start: file offset of the range start + * @end: file offset of the range end (inclusive) + * + * Return a bitmap where bit set means a csum mismatch, and bit not set means + * csum match. */ -static int btrfs_readpage_end_io_hook(struct btrfs_io_bio *io_bio, - u64 phy_offset, struct page *page, - u64 start, u64 end, int mirror) +unsigned int btrfs_verify_data_csum(struct btrfs_bio *bbio, + u32 bio_offset, struct page *page, + u64 start, u64 end) { - size_t offset = start - page_offset(page); struct inode *inode = page->mapping->host; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_root *root = BTRFS_I(inode)->root; + const u32 sectorsize = root->fs_info->sectorsize; + u32 pg_off; + unsigned int result = 0; - if (PageChecked(page)) { - ClearPageChecked(page); + /* + * This only happens for NODATASUM or compressed read. + * Normally this should be covered by above check for compressed read + * or the next check for NODATASUM. Just do a quicker exit here. + */ + if (bbio->csum == NULL) return 0; - } if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) return 0; - if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID && - test_range_bit(io_tree, start, end, EXTENT_NODATASUM, 1, NULL)) { - clear_extent_bits(io_tree, start, end, EXTENT_NODATASUM); + if (unlikely(test_bit(BTRFS_FS_STATE_NO_CSUMS, &fs_info->fs_state))) return 0; - } - phy_offset >>= inode->i_sb->s_blocksize_bits; - return __readpage_endio_check(inode, io_bio, phy_offset, page, offset, - start, (size_t)(end - start + 1)); + ASSERT(page_offset(page) <= start && + end <= page_offset(page) + PAGE_SIZE - 1); + for (pg_off = offset_in_page(start); + pg_off < offset_in_page(end); + pg_off += sectorsize, bio_offset += sectorsize) { + u64 file_offset = pg_off + page_offset(page); + int ret; + + if (btrfs_is_data_reloc_root(root) && + test_range_bit(io_tree, file_offset, + file_offset + sectorsize - 1, + EXTENT_NODATASUM, 1, NULL)) { + /* Skip the range without csum for data reloc inode */ + clear_extent_bits(io_tree, file_offset, + file_offset + sectorsize - 1, + EXTENT_NODATASUM); + continue; + } + ret = btrfs_check_data_csum(inode, bbio, bio_offset, page, pg_off); + if (ret < 0) { + const int nr_bit = (pg_off - offset_in_page(start)) >> + root->fs_info->sectorsize_bits; + + result |= (1U << nr_bit); + } + } + return result; } /* @@ -2838,19 +3612,22 @@ void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info) inode = list_first_entry(&fs_info->delayed_iputs, struct btrfs_inode, delayed_iput); run_delayed_iput_locked(fs_info, inode); + cond_resched_lock(&fs_info->delayed_iput_lock); } spin_unlock(&fs_info->delayed_iput_lock); } /** - * btrfs_wait_on_delayed_iputs - wait on the delayed iputs to be done running - * @fs_info - the fs_info for this fs - * @return - EINTR if we were killed, 0 if nothing's pending + * Wait for flushing all delayed iputs + * + * @fs_info: the filesystem * * This will wait on any delayed iputs that are currently running with KILLABLE * set. Once they are all done running we will return, unless we are killed in * which case we return EINTR. This helps in user operations like fallocate etc * that might get blocked on the iputs. + * + * Return EINTR if we were killed, 0 if nothing's pending */ int btrfs_wait_on_delayed_iputs(struct btrfs_fs_info *fs_info) { @@ -2904,7 +3681,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) u64 last_objectid = 0; int ret = 0, nr_unlink = 0; - if (cmpxchg(&root->orphan_cleanup_state, 0, ORPHAN_CLEANUP_STARTED)) + if (test_and_set_bit(BTRFS_ROOT_ORPHAN_CLEANUP, &root->state)) return 0; path = btrfs_alloc_path(); @@ -2966,37 +3743,38 @@ 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(fs_info->sb, &found_key, root); + inode = btrfs_iget(fs_info->sb, last_objectid, root); ret = PTR_ERR_OR_ZERO(inode); if (ret && ret != -ENOENT) goto out; 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; /* - * this is an orphan in the tree root. Currently these + * This is an orphan in the tree root. Currently these * could come from 2 sources: - * a) a snapshot deletion in progress + * a) a root (snapshot/subvolume) deletion in progress * b) a free space cache inode - * We need to distinguish those two, as the snapshot - * orphan must not get deleted. - * find_dead_roots already ran before us, so if this - * is a snapshot deletion, we should find the root - * in the dead_roots list + * We need to distinguish those two, as the orphan item + * for a root must not get deleted before the deletion + * of the snapshot/subvolume's tree completes. + * + * btrfs_find_orphan_roots() ran before us, which has + * found all deleted roots and loaded them into + * fs_info->fs_roots_radix. So here we can find if an + * orphan item corresponds to a deleted root by looking + * up the root from that radix tree. */ - spin_lock(&fs_info->trans_lock); - list_for_each_entry(dead_root, &fs_info->dead_roots, - root_list) { - if (dead_root->root_key.objectid == - found_key.objectid) { - is_dead_root = 1; - break; - } - } - spin_unlock(&fs_info->trans_lock); + + spin_lock(&fs_info->fs_roots_radix_lock); + dead_root = radix_tree_lookup(&fs_info->fs_roots_radix, + (unsigned long)found_key.objectid); + if (dead_root && btrfs_root_refs(&dead_root->root_item) == 0) + is_dead_root = 1; + spin_unlock(&fs_info->fs_roots_radix_lock); + if (is_dead_root) { /* prevent this orphan from being found again */ key.offset = found_key.objectid - 1; @@ -3007,7 +3785,14 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) /* * If we have an inode with links, there are a couple of - * possibilities. Old kernels (before v3.12) used to create an + * possibilities: + * + * 1. We were halfway through creating fsverity metadata for the + * file. In that case, the orphan item represents incomplete + * fsverity metadata which must be cleaned up with + * btrfs_drop_verity_items and deleting the orphan item. + + * 2. Old kernels (before v3.12) used to create an * orphan item for truncate indicating that there were possibly * extent items past i_size that needed to be deleted. In v3.12, * truncate was changed to update i_size in sync with the extent @@ -3025,8 +3810,12 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) * but either way, we can delete the orphan item. */ if (ret == -ENOENT || inode->i_nlink) { - if (!ret) + if (!ret) { + ret = btrfs_drop_verity_items(BTRFS_I(inode)); iput(inode); + if (ret) + goto out; + } trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { ret = PTR_ERR(trans); @@ -3050,8 +3839,6 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) /* release the path since we're done with it */ btrfs_release_path(path); - root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE; - if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) { trans = btrfs_join_transaction(root); if (!IS_ERR(trans)) @@ -3187,6 +3974,8 @@ static int btrfs_read_locked_inode(struct inode *inode, i_uid_write(inode, btrfs_inode_uid(leaf, inode_item)); i_gid_write(inode, btrfs_inode_gid(leaf, inode_item)); btrfs_i_size_write(BTRFS_I(inode), btrfs_inode_size(leaf, inode_item)); + btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0, + round_up(i_size_read(inode), fs_info->sectorsize)); inode->i_atime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->atime); inode->i_atime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->atime); @@ -3213,7 +4002,8 @@ static int btrfs_read_locked_inode(struct inode *inode, rdev = btrfs_inode_rdev(leaf, inode_item); BTRFS_I(inode)->index_cnt = (u64)-1; - BTRFS_I(inode)->flags = btrfs_inode_flags(leaf, inode_item); + btrfs_inode_split_flags(btrfs_inode_flags(leaf, inode_item), + &BTRFS_I(inode)->flags, &BTRFS_I(inode)->ro_flags); cache_index: /* @@ -3258,6 +4048,14 @@ cache_index: */ BTRFS_I(inode)->last_unlink_trans = BTRFS_I(inode)->last_trans; + /* + * Same logic as for last_unlink_trans. We don't persist the generation + * of the last transaction where this inode was used for a reflink + * operation, so after eviction and reloading the inode we must be + * pessimistic and assume the last transaction that modified the inode. + */ + BTRFS_I(inode)->last_reflink_trans = BTRFS_I(inode)->last_trans; + path->slots[0]++; if (inode->i_nlink != 1 || path->slots[0] >= btrfs_header_nritems(leaf)) @@ -3305,7 +4103,6 @@ cache_acl: switch (inode->i_mode & S_IFMT) { case S_IFREG: inode->i_mapping->a_ops = &btrfs_aops; - BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; inode->i_fop = &btrfs_file_operations; inode->i_op = &btrfs_file_inode_operations; break; @@ -3337,53 +4134,54 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, struct inode *inode) { struct btrfs_map_token token; + u64 flags; btrfs_init_map_token(&token, leaf); - btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token); - btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token); - btrfs_set_token_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size, - &token); - btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token); - btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token); - - btrfs_set_token_timespec_sec(leaf, &item->atime, - inode->i_atime.tv_sec, &token); - btrfs_set_token_timespec_nsec(leaf, &item->atime, - inode->i_atime.tv_nsec, &token); - - btrfs_set_token_timespec_sec(leaf, &item->mtime, - inode->i_mtime.tv_sec, &token); - btrfs_set_token_timespec_nsec(leaf, &item->mtime, - inode->i_mtime.tv_nsec, &token); - - btrfs_set_token_timespec_sec(leaf, &item->ctime, - inode->i_ctime.tv_sec, &token); - btrfs_set_token_timespec_nsec(leaf, &item->ctime, - inode->i_ctime.tv_nsec, &token); - - btrfs_set_token_timespec_sec(leaf, &item->otime, - BTRFS_I(inode)->i_otime.tv_sec, &token); - btrfs_set_token_timespec_nsec(leaf, &item->otime, - BTRFS_I(inode)->i_otime.tv_nsec, &token); - - btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode), - &token); - btrfs_set_token_inode_generation(leaf, item, BTRFS_I(inode)->generation, - &token); - btrfs_set_token_inode_sequence(leaf, item, inode_peek_iversion(inode), - &token); - btrfs_set_token_inode_transid(leaf, item, trans->transid, &token); - btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token); - btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token); - btrfs_set_token_inode_block_group(leaf, item, 0, &token); + btrfs_set_token_inode_uid(&token, item, i_uid_read(inode)); + btrfs_set_token_inode_gid(&token, item, i_gid_read(inode)); + btrfs_set_token_inode_size(&token, item, BTRFS_I(inode)->disk_i_size); + btrfs_set_token_inode_mode(&token, item, inode->i_mode); + btrfs_set_token_inode_nlink(&token, item, inode->i_nlink); + + btrfs_set_token_timespec_sec(&token, &item->atime, + inode->i_atime.tv_sec); + btrfs_set_token_timespec_nsec(&token, &item->atime, + inode->i_atime.tv_nsec); + + btrfs_set_token_timespec_sec(&token, &item->mtime, + inode->i_mtime.tv_sec); + btrfs_set_token_timespec_nsec(&token, &item->mtime, + inode->i_mtime.tv_nsec); + + btrfs_set_token_timespec_sec(&token, &item->ctime, + inode->i_ctime.tv_sec); + btrfs_set_token_timespec_nsec(&token, &item->ctime, + inode->i_ctime.tv_nsec); + + btrfs_set_token_timespec_sec(&token, &item->otime, + BTRFS_I(inode)->i_otime.tv_sec); + btrfs_set_token_timespec_nsec(&token, &item->otime, + BTRFS_I(inode)->i_otime.tv_nsec); + + btrfs_set_token_inode_nbytes(&token, item, inode_get_bytes(inode)); + btrfs_set_token_inode_generation(&token, item, + BTRFS_I(inode)->generation); + btrfs_set_token_inode_sequence(&token, item, inode_peek_iversion(inode)); + btrfs_set_token_inode_transid(&token, item, trans->transid); + btrfs_set_token_inode_rdev(&token, item, inode->i_rdev); + flags = btrfs_inode_combine_flags(BTRFS_I(inode)->flags, + BTRFS_I(inode)->ro_flags); + btrfs_set_token_inode_flags(&token, item, flags); + btrfs_set_token_inode_block_group(&token, item, 0); } /* * copy everything in the in-memory inode into the btree. */ static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans, - struct btrfs_root *root, struct inode *inode) + struct btrfs_root *root, + struct btrfs_inode *inode) { struct btrfs_inode_item *inode_item; struct btrfs_path *path; @@ -3394,9 +4192,7 @@ static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; - path->leave_spinning = 1; - ret = btrfs_lookup_inode(trans, root, path, &BTRFS_I(inode)->location, - 1); + ret = btrfs_lookup_inode(trans, root, path, &inode->location, 1); if (ret) { if (ret > 0) ret = -ENOENT; @@ -3407,7 +4203,7 @@ static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans, inode_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item); - fill_inode_item(trans, leaf, inode_item, inode); + fill_inode_item(trans, leaf, inode_item, &inode->vfs_inode); btrfs_mark_buffer_dirty(leaf); btrfs_set_inode_last_trans(trans, inode); ret = 0; @@ -3420,7 +4216,8 @@ failed: * copy everything in the in-memory inode into the btree. */ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, - struct btrfs_root *root, struct inode *inode) + struct btrfs_root *root, + struct btrfs_inode *inode) { struct btrfs_fs_info *fs_info = root->fs_info; int ret; @@ -3432,8 +4229,8 @@ 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(BTRFS_I(inode)) - && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID + if (!btrfs_is_free_space_inode(inode) + && !btrfs_is_data_reloc_root(root) && !test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) { btrfs_update_root_times(trans, root); @@ -3446,9 +4243,8 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, return btrfs_update_inode_item(trans, root, inode); } -noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct inode *inode) +int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct btrfs_inode *inode) { int ret; @@ -3464,11 +4260,12 @@ noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, * also drops the back refs in the inode to the directory */ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, - struct btrfs_root *root, struct btrfs_inode *dir, struct btrfs_inode *inode, - const char *name, int name_len) + const char *name, int name_len, + struct btrfs_rename_ctx *rename_ctx) { + struct btrfs_root *root = dir->root; struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; int ret = 0; @@ -3483,7 +4280,6 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, goto out; } - path->leave_spinning = 1; di = btrfs_lookup_dir_item(trans, root, path, dir_ino, name, name_len, -1); if (IS_ERR_OR_NULL(di)) { @@ -3523,26 +4319,28 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, goto err; } skip_backref: + if (rename_ctx) + rename_ctx->index = index; + ret = btrfs_delete_delayed_dir_index(trans, 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); - if (ret != 0 && ret != -ENOENT) { - btrfs_abort_transaction(trans, ret); - goto err; + /* + * If we are in a rename context, we don't need to update anything in the + * log. That will be done later during the rename by btrfs_log_new_name(). + * Besides that, doing it here would only cause extra unnecessary btree + * operations on the log tree, increasing latency for applications. + */ + if (!rename_ctx) { + btrfs_del_inode_ref_in_log(trans, root, name, name_len, inode, + dir_ino); + 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) - btrfs_abort_transaction(trans, ret); - /* * If we have a pending delayed iput we could end up with the final iput * being run in btrfs-cleaner context. If we have enough of these built @@ -3561,23 +4359,23 @@ err: 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); + inode->vfs_inode.i_ctime = current_time(&inode->vfs_inode); + dir->vfs_inode.i_mtime = inode->vfs_inode.i_ctime; + dir->vfs_inode.i_ctime = inode->vfs_inode.i_ctime; + ret = btrfs_update_inode(trans, root, dir); out: return ret; } int btrfs_unlink_inode(struct btrfs_trans_handle *trans, - struct btrfs_root *root, 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); + ret = __btrfs_unlink_inode(trans, dir, inode, name, name_len, NULL); if (!ret) { drop_nlink(&inode->vfs_inode); - ret = btrfs_update_inode(trans, root, &inode->vfs_inode); + ret = btrfs_update_inode(trans, inode->root, inode); } return ret; } @@ -3600,13 +4398,13 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir) * 1 for the dir index * 1 for the inode ref * 1 for the inode + * 1 for the parent inode */ - return btrfs_start_transaction_fallback_global_rsv(root, 5, 5); + return btrfs_start_transaction_fallback_global_rsv(root, 6); } static int btrfs_unlink(struct inode *dir, struct dentry *dentry) { - struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_trans_handle *trans; struct inode *inode = d_inode(dentry); int ret; @@ -3618,7 +4416,7 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry) btrfs_record_unlink_dir(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), 0); - ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir), + ret = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), dentry->d_name.name, dentry->d_name.len); if (ret) @@ -3632,7 +4430,7 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry) out: btrfs_end_transaction(trans); - btrfs_btree_balance_dirty(root->fs_info); + btrfs_btree_balance_dirty(BTRFS_I(dir)->root->fs_info); return ret; } @@ -3686,7 +4484,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, * This is a placeholder inode for a subvolume we didn't have a * reference to at the time of the snapshot creation. In the meantime * we could have renamed the real subvol link into our snapshot, so - * depending on btrfs_del_root_ref to return -ENOENT here is incorret. + * depending on btrfs_del_root_ref to return -ENOENT here is incorrect. * Instead simply lookup the dir_index_item for this entry so we can * remove it. Otherwise we know we have a ref to the root and we can * call btrfs_del_root_ref, and it _shouldn't_ fail. @@ -3725,8 +4523,9 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, 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); + dir->i_mtime = current_time(dir); + dir->i_ctime = dir->i_mtime; + ret = btrfs_update_inode_fallback(trans, root, BTRFS_I(dir)); if (ret) btrfs_abort_transaction(trans, ret); out: @@ -3799,7 +4598,7 @@ static void btrfs_prune_dentries(struct btrfs_root *root) struct inode *inode; u64 objectid = 0; - if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + if (!BTRFS_FS_ERROR(fs_info)) WARN_ON(btrfs_root_refs(&root->root_item) != 0); spin_lock(&root->inode_lock); @@ -3863,7 +4662,6 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) struct btrfs_block_rsv block_rsv; u64 root_flags; int ret; - int err; /* * Don't allow to delete a subvolume with send in progress. This is @@ -3878,6 +4676,13 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) dest->root_key.objectid); return -EPERM; } + if (atomic_read(&dest->nr_swapfiles)) { + spin_unlock(&dest->root_item_lock); + btrfs_warn(fs_info, + "attempt to delete subvolume %llu with active swapfile", + root->root_key.objectid); + return -EPERM; + } root_flags = btrfs_root_flags(&dest->root_item); btrfs_set_root_flags(&dest->root_item, root_flags | BTRFS_ROOT_SUBVOL_DEAD); @@ -3885,8 +4690,8 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) down_write(&fs_info->subvol_sem); - err = may_destroy_subvol(dest); - if (err) + ret = may_destroy_subvol(dest); + if (ret) goto out_up_write; btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); @@ -3895,13 +4700,13 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) * two for dir entries, * two for root ref/backref. */ - err = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true); - if (err) + ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true); + if (ret) goto out_up_write; trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { - err = PTR_ERR(trans); + ret = PTR_ERR(trans); goto out_release; } trans->block_rsv = &block_rsv; @@ -3911,16 +4716,19 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) ret = btrfs_unlink_subvol(trans, dir, dentry); if (ret) { - err = ret; btrfs_abort_transaction(trans, ret); goto out_end_trans; } - btrfs_record_root_in_trans(trans, dest); + ret = btrfs_record_root_in_trans(trans, dest); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_end_trans; + } memset(&dest->root_item.drop_progress, 0, sizeof(dest->root_item.drop_progress)); - dest->root_item.drop_level = 0; + btrfs_set_root_drop_level(&dest->root_item, 0); btrfs_set_root_refs(&dest->root_item, 0); if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) { @@ -3929,7 +4737,6 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) dest->root_key.objectid); if (ret) { btrfs_abort_transaction(trans, ret); - err = ret; goto out_end_trans; } } @@ -3939,7 +4746,6 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) dest->root_key.objectid); if (ret && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); - err = ret; goto out_end_trans; } if (!btrfs_is_empty_uuid(dest->root_item.received_uuid)) { @@ -3949,23 +4755,22 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) dest->root_key.objectid); if (ret && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); - err = ret; goto out_end_trans; } } + free_anon_bdev(dest->anon_dev); + dest->anon_dev = 0; out_end_trans: trans->block_rsv = NULL; trans->bytes_reserved = 0; ret = btrfs_end_transaction(trans); - if (ret && !err) - err = ret; inode->i_flags |= S_DEAD; out_release: - btrfs_subvolume_release_metadata(fs_info, &block_rsv); + btrfs_subvolume_release_metadata(root, &block_rsv); out_up_write: up_write(&fs_info->subvol_sem); - if (err) { + if (ret) { spin_lock(&dest->root_item_lock); root_flags = btrfs_root_flags(&dest->root_item); btrfs_set_root_flags(&dest->root_item, @@ -3975,29 +4780,29 @@ out_up_write: d_invalidate(dentry); btrfs_prune_dentries(dest); ASSERT(dest->send_in_progress == 0); - - /* the last ref */ - if (dest->ino_cache_inode) { - iput(dest->ino_cache_inode); - dest->ino_cache_inode = NULL; - } } - return err; + return ret; } static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) { struct inode *inode = d_inode(dentry); + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; int err = 0; - struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_trans_handle *trans; u64 last_unlink_trans; if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) return -ENOTEMPTY; - if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID) + if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID) { + if (unlikely(btrfs_fs_incompat(fs_info, EXTENT_TREE_V2))) { + btrfs_err(fs_info, + "extent tree v2 doesn't support snapshot deletion yet"); + return -EOPNOTSUPP; + } return btrfs_delete_subvolume(dir, dentry); + } trans = __unlink_start_trans(dir); if (IS_ERR(trans)) @@ -4015,7 +4820,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; /* now the directory is empty */ - err = btrfs_unlink_inode(trans, root, BTRFS_I(dir), + err = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), dentry->d_name.name, dentry->d_name.len); if (!err) { @@ -4036,348 +4841,12 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) } out: btrfs_end_transaction(trans); - btrfs_btree_balance_dirty(root->fs_info); + btrfs_btree_balance_dirty(fs_info); return err; } /* - * Return this if we need to call truncate_block for the last bit of the - * truncate. - */ -#define NEED_TRUNCATE_BLOCK 1 - -/* - * this can truncate away extent items, csum items and directory items. - * It starts at a high offset and removes keys until it can't find - * any higher than new_size - * - * csum items that cross the new i_size are truncated to the new size - * as well. - * - * min_type is the minimum key type to truncate down to. If set to 0, this - * will kill all the items on this inode, including the INODE_ITEM_KEY. - */ -int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - 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; - struct btrfs_key key; - struct btrfs_key found_key; - u64 extent_start = 0; - u64 extent_num_bytes = 0; - u64 extent_offset = 0; - u64 item_end = 0; - u64 last_size = new_size; - u32 found_type = (u8)-1; - int found_extent; - int del_item; - int pending_del_nr = 0; - int pending_del_slot = 0; - int extent_type = -1; - int ret; - u64 ino = btrfs_ino(BTRFS_I(inode)); - u64 bytes_deleted = 0; - bool be_nice = false; - bool should_throttle = false; - const u64 lock_start = ALIGN_DOWN(new_size, fs_info->sectorsize); - struct extent_state *cached_state = NULL; - - BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY); - - /* - * for non-free space inodes and ref cows, we want to back off from - * time to time - */ - if (!btrfs_is_free_space_inode(BTRFS_I(inode)) && - test_bit(BTRFS_ROOT_REF_COWS, &root->state)) - be_nice = true; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - path->reada = READA_BACK; - - if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) - lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, - &cached_state); - - /* - * We want to drop from the next block forward in case this new size is - * not block aligned since we will be keeping the last block of the - * extent just the way it is. - */ - if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) || - 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 - * we relog the inode, so if root != BTRFS_I(inode)->root, it means - * it is used to drop the logged items. So we shouldn't kill the delayed - * items. - */ - if (min_type == 0 && root == BTRFS_I(inode)->root) - btrfs_kill_delayed_inode_items(BTRFS_I(inode)); - - key.objectid = ino; - key.offset = (u64)-1; - key.type = (u8)-1; - -search_again: - /* - * with a 16K leaf size and 128MB extents, you can actually queue - * up a huge file in a single leaf. Most of the time that - * bytes_deleted is > 0, it will be huge by the time we get here - */ - if (be_nice && bytes_deleted > SZ_32M && - btrfs_should_end_transaction(trans)) { - ret = -EAGAIN; - goto out; - } - - ret = btrfs_search_slot(trans, root, &key, path, -1, 1); - if (ret < 0) - goto out; - - if (ret > 0) { - ret = 0; - /* there are no items in the tree for us to truncate, we're - * done - */ - if (path->slots[0] == 0) - goto out; - path->slots[0]--; - } - - while (1) { - fi = NULL; - leaf = path->nodes[0]; - btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); - found_type = found_key.type; - - if (found_key.objectid != ino) - break; - - if (found_type < min_type) - break; - - item_end = found_key.offset; - if (found_type == BTRFS_EXTENT_DATA_KEY) { - fi = btrfs_item_ptr(leaf, path->slots[0], - struct btrfs_file_extent_item); - extent_type = btrfs_file_extent_type(leaf, fi); - if (extent_type != BTRFS_FILE_EXTENT_INLINE) { - item_end += - btrfs_file_extent_num_bytes(leaf, fi); - - trace_btrfs_truncate_show_fi_regular( - BTRFS_I(inode), leaf, fi, - found_key.offset); - } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - item_end += btrfs_file_extent_ram_bytes(leaf, - fi); - - trace_btrfs_truncate_show_fi_inline( - BTRFS_I(inode), leaf, fi, path->slots[0], - found_key.offset); - } - item_end--; - } - if (found_type > min_type) { - del_item = 1; - } else { - if (item_end < new_size) - break; - if (found_key.offset >= new_size) - del_item = 1; - else - del_item = 0; - } - found_extent = 0; - /* FIXME, shrink the extent if the ref count is only 1 */ - if (found_type != BTRFS_EXTENT_DATA_KEY) - goto delete; - - if (extent_type != BTRFS_FILE_EXTENT_INLINE) { - u64 num_dec; - extent_start = btrfs_file_extent_disk_bytenr(leaf, fi); - if (!del_item) { - u64 orig_num_bytes = - btrfs_file_extent_num_bytes(leaf, fi); - extent_num_bytes = ALIGN(new_size - - found_key.offset, - fs_info->sectorsize); - btrfs_set_file_extent_num_bytes(leaf, fi, - extent_num_bytes); - num_dec = (orig_num_bytes - - extent_num_bytes); - if (test_bit(BTRFS_ROOT_REF_COWS, - &root->state) && - extent_start != 0) - inode_sub_bytes(inode, num_dec); - btrfs_mark_buffer_dirty(leaf); - } else { - extent_num_bytes = - btrfs_file_extent_disk_num_bytes(leaf, - fi); - extent_offset = found_key.offset - - btrfs_file_extent_offset(leaf, fi); - - /* FIXME blocksize != 4096 */ - num_dec = btrfs_file_extent_num_bytes(leaf, fi); - if (extent_start != 0) { - found_extent = 1; - if (test_bit(BTRFS_ROOT_REF_COWS, - &root->state)) - inode_sub_bytes(inode, num_dec); - } - } - } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - /* - * we can't truncate inline items that have had - * special encodings - */ - if (!del_item && - btrfs_file_extent_encryption(leaf, fi) == 0 && - btrfs_file_extent_other_encoding(leaf, fi) == 0 && - btrfs_file_extent_compression(leaf, fi) == 0) { - u32 size = (u32)(new_size - found_key.offset); - - btrfs_set_file_extent_ram_bytes(leaf, fi, size); - size = btrfs_file_extent_calc_inline_size(size); - btrfs_truncate_item(path, size, 1); - } else if (!del_item) { - /* - * We have to bail so the last_size is set to - * just before this extent. - */ - ret = NEED_TRUNCATE_BLOCK; - break; - } - - if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) - inode_sub_bytes(inode, item_end + 1 - new_size); - } -delete: - if (del_item) - last_size = found_key.offset; - else - last_size = new_size; - if (del_item) { - if (!pending_del_nr) { - /* no pending yet, add ourselves */ - pending_del_slot = path->slots[0]; - pending_del_nr = 1; - } else if (pending_del_nr && - path->slots[0] + 1 == pending_del_slot) { - /* hop on the pending chunk */ - pending_del_nr++; - pending_del_slot = path->slots[0]; - } else { - BUG(); - } - } else { - break; - } - should_throttle = false; - - if (found_extent && - (test_bit(BTRFS_ROOT_REF_COWS, &root->state) || - root == fs_info->tree_root)) { - struct btrfs_ref ref = { 0 }; - - bytes_deleted += extent_num_bytes; - - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, - extent_start, extent_num_bytes, 0); - ref.real_root = root->root_key.objectid; - btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), - ino, extent_offset); - ret = btrfs_free_extent(trans, &ref); - if (ret) { - btrfs_abort_transaction(trans, ret); - break; - } - if (be_nice) { - if (btrfs_should_throttle_delayed_refs(trans)) - should_throttle = true; - } - } - - if (found_type == BTRFS_INODE_ITEM_KEY) - break; - - if (path->slots[0] == 0 || - path->slots[0] != pending_del_slot || - should_throttle) { - if (pending_del_nr) { - ret = btrfs_del_items(trans, root, path, - pending_del_slot, - pending_del_nr); - if (ret) { - btrfs_abort_transaction(trans, ret); - break; - } - pending_del_nr = 0; - } - btrfs_release_path(path); - - /* - * We can generate a lot of delayed refs, so we need to - * throttle every once and a while and make sure we're - * adding enough space to keep up with the work we are - * generating. Since we hold a transaction here we - * can't flush, and we don't want to FLUSH_LIMIT because - * we could have generated too many delayed refs to - * actually allocate, so just bail if we're short and - * let the normal reservation dance happen higher up. - */ - if (should_throttle) { - ret = btrfs_delayed_refs_rsv_refill(fs_info, - BTRFS_RESERVE_NO_FLUSH); - if (ret) { - ret = -EAGAIN; - break; - } - } - goto search_again; - } else { - path->slots[0]--; - } - } -out: - if (ret >= 0 && pending_del_nr) { - int err; - - err = btrfs_del_items(trans, root, path, pending_del_slot, - pending_del_nr); - if (err) { - btrfs_abort_transaction(trans, err); - ret = err; - } - } - if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { - ASSERT(last_size >= new_size); - if (!ret && last_size > new_size) - last_size = new_size; - btrfs_ordered_update_i_size(inode, last_size, NULL); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, - (u64)-1, &cached_state); - } - - btrfs_free_path(path); - return ret; -} - -/* * btrfs_truncate_block - read, zero a chunk and write a block * @inode - inode that we're zeroing * @from - the offset to start zeroing @@ -4388,21 +4857,22 @@ out: * This will find the block for the "from" offset and cow the block and zero the * part we want to zero. This is used with truncate and hole punching. */ -int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, - int front) +int btrfs_truncate_block(struct btrfs_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 extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct address_space *mapping = inode->vfs_inode.i_mapping; + struct extent_io_tree *io_tree = &inode->io_tree; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; struct extent_changeset *data_reserved = NULL; - char *kaddr; + bool only_release_metadata = false; u32 blocksize = fs_info->sectorsize; pgoff_t index = from >> PAGE_SHIFT; unsigned offset = from & (blocksize - 1); struct page *page; gfp_t mask = btrfs_alloc_write_mask(mapping); + size_t write_bytes = blocksize; int ret = 0; u64 block_start; u64 block_end; @@ -4414,23 +4884,38 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, block_start = round_down(from, blocksize); block_end = block_start + blocksize - 1; - ret = btrfs_delalloc_reserve_space(inode, &data_reserved, - block_start, blocksize); - if (ret) + ret = btrfs_check_data_free_space(inode, &data_reserved, block_start, + blocksize, false); + if (ret < 0) { + if (btrfs_check_nocow_lock(inode, block_start, &write_bytes, false) > 0) { + /* For nocow case, no need to reserve data space */ + only_release_metadata = true; + } else { + goto out; + } + } + ret = btrfs_delalloc_reserve_metadata(inode, blocksize, blocksize, false); + if (ret < 0) { + if (!only_release_metadata) + btrfs_free_reserved_data_space(inode, data_reserved, + block_start, blocksize); goto out; - + } again: page = find_or_create_page(mapping, index, mask); if (!page) { - btrfs_delalloc_release_space(inode, data_reserved, - block_start, blocksize, true); - btrfs_delalloc_release_extents(BTRFS_I(inode), blocksize); + btrfs_delalloc_release_space(inode, data_reserved, block_start, + blocksize, true); + btrfs_delalloc_release_extents(inode, blocksize); ret = -ENOMEM; goto out; } + ret = set_page_extent_mapped(page); + if (ret < 0) + goto out_unlock; if (!PageUptodate(page)) { - ret = btrfs_readpage(NULL, page); + ret = btrfs_read_folio(NULL, page_folio(page)); lock_page(page); if (page->mapping != mapping) { unlock_page(page); @@ -4444,78 +4929,82 @@ again: } wait_on_page_writeback(page); - lock_extent_bits(io_tree, block_start, block_end, &cached_state); - set_page_extent_mapped(page); + lock_extent(io_tree, block_start, block_end, &cached_state); ordered = btrfs_lookup_ordered_extent(inode, block_start); if (ordered) { - unlock_extent_cached(io_tree, block_start, block_end, - &cached_state); + unlock_extent(io_tree, block_start, block_end, &cached_state); unlock_page(page); put_page(page); - btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_start_ordered_extent(ordered, 1); btrfs_put_ordered_extent(ordered); goto again; } - clear_extent_bit(&BTRFS_I(inode)->io_tree, block_start, block_end, + clear_extent_bit(&inode->io_tree, block_start, block_end, EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, - 0, 0, &cached_state); + &cached_state); ret = btrfs_set_extent_delalloc(inode, block_start, block_end, 0, &cached_state); if (ret) { - unlock_extent_cached(io_tree, block_start, block_end, - &cached_state); + unlock_extent(io_tree, block_start, block_end, &cached_state); goto out_unlock; } if (offset != blocksize) { if (!len) len = blocksize - offset; - kaddr = kmap(page); if (front) - memset(kaddr + (block_start - page_offset(page)), - 0, offset); + memzero_page(page, (block_start - page_offset(page)), + offset); else - memset(kaddr + (block_start - page_offset(page)) + offset, - 0, len); - flush_dcache_page(page); - kunmap(page); + memzero_page(page, (block_start - page_offset(page)) + offset, + len); } - ClearPageChecked(page); - set_page_dirty(page); - unlock_extent_cached(io_tree, block_start, block_end, &cached_state); + btrfs_page_clear_checked(fs_info, page, block_start, + block_end + 1 - block_start); + btrfs_page_set_dirty(fs_info, page, block_start, block_end + 1 - block_start); + unlock_extent(io_tree, block_start, block_end, &cached_state); + + if (only_release_metadata) + set_extent_bit(&inode->io_tree, block_start, block_end, + EXTENT_NORESERVE, NULL, GFP_NOFS); out_unlock: - if (ret) - btrfs_delalloc_release_space(inode, data_reserved, block_start, - blocksize, true); - btrfs_delalloc_release_extents(BTRFS_I(inode), blocksize); + if (ret) { + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, blocksize, true); + else + btrfs_delalloc_release_space(inode, data_reserved, + block_start, blocksize, true); + } + btrfs_delalloc_release_extents(inode, blocksize); unlock_page(page); put_page(page); out: + if (only_release_metadata) + btrfs_check_nocow_unlock(inode); extent_changeset_free(data_reserved); return ret; } -static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, +static int maybe_insert_hole(struct btrfs_root *root, struct btrfs_inode *inode, u64 offset, u64 len) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_trans_handle *trans; + struct btrfs_drop_extents_args drop_args = { 0 }; int ret; /* - * Still need to make sure the inode looks like it's been updated so - * that any holes get logged if we fsync. + * If NO_HOLES is enabled, we don't need to do anything. + * Later, up in the call chain, either btrfs_set_inode_last_sub_trans() + * or btrfs_update_inode() will be called, which guarantee that the next + * fsync will know this inode was changed and needs to be logged. */ - 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; + if (btrfs_fs_incompat(fs_info, NO_HOLES)) return 0; - } /* * 1 - for the one we're dropping @@ -4526,19 +5015,24 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, if (IS_ERR(trans)) return PTR_ERR(trans); - ret = btrfs_drop_extents(trans, root, inode, offset, offset + len, 1); + drop_args.start = offset; + drop_args.end = offset + len; + drop_args.drop_cache = true; + + ret = btrfs_drop_extents(trans, root, inode, &drop_args); if (ret) { btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); return ret; } - ret = btrfs_insert_file_extent(trans, root, btrfs_ino(BTRFS_I(inode)), - offset, 0, 0, len, 0, len, 0, 0, 0); - if (ret) + ret = btrfs_insert_hole_extent(trans, root, btrfs_ino(inode), offset, len); + if (ret) { btrfs_abort_transaction(trans, ret); - else + } else { + btrfs_update_inode_bytes(inode, 0, drop_args.bytes_found); btrfs_update_inode(trans, root, inode); + } btrfs_end_transaction(trans); return ret; } @@ -4549,14 +5043,13 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode, * these file extents so that btrfs_get_extent will return a EXTENT_MAP_HOLE for * the range between oldsize and size */ -int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) +int btrfs_cont_expand(struct btrfs_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 btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_io_tree *io_tree = &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, fs_info->sectorsize); u64 block_end = ALIGN(size, fs_info->sectorsize); u64 last_byte; @@ -4576,11 +5069,11 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) if (size <= hole_start) return 0; - btrfs_lock_and_flush_ordered_range(io_tree, BTRFS_I(inode), hole_start, - block_end - 1, &cached_state); + btrfs_lock_and_flush_ordered_range(inode, hole_start, block_end - 1, + &cached_state); cur_offset = hole_start; while (1) { - em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset, + em = btrfs_get_extent(inode, NULL, 0, cur_offset, block_end - cur_offset); if (IS_ERR(em)) { err = PTR_ERR(em); @@ -4589,20 +5082,27 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) } last_byte = min(extent_map_end(em), block_end); last_byte = ALIGN(last_byte, fs_info->sectorsize); + hole_size = last_byte - cur_offset; + if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { struct extent_map *hole_em; - hole_size = last_byte - cur_offset; err = maybe_insert_hole(root, inode, cur_offset, hole_size); if (err) break; - btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, - cur_offset + hole_size - 1, 0); + + err = btrfs_inode_set_file_extent_range(inode, + cur_offset, hole_size); + if (err) + break; + hole_em = alloc_extent_map(); if (!hole_em) { - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &BTRFS_I(inode)->runtime_flags); + btrfs_drop_extent_map_range(inode, cur_offset, + cur_offset + hole_size - 1, + false); + btrfs_set_inode_full_sync(inode); goto next; } hole_em->start = cur_offset; @@ -4616,18 +5116,13 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) hole_em->compress_type = BTRFS_COMPRESS_NONE; hole_em->generation = fs_info->generation; - while (1) { - write_lock(&em_tree->lock); - err = add_extent_mapping(em_tree, hole_em, 1); - write_unlock(&em_tree->lock); - if (err != -EEXIST) - break; - btrfs_drop_extent_cache(BTRFS_I(inode), - cur_offset, - cur_offset + - hole_size - 1, 0); - } + err = btrfs_replace_extent_map_range(inode, hole_em, true); free_extent_map(hole_em); + } else { + err = btrfs_inode_set_file_extent_range(inode, + cur_offset, hole_size); + if (err) + break; } next: free_extent_map(em); @@ -4637,7 +5132,7 @@ next: break; } free_extent_map(em); - unlock_extent_cached(io_tree, hole_start, block_end - 1, &cached_state); + unlock_extent(io_tree, hole_start, block_end - 1, &cached_state); return err; } @@ -4658,9 +5153,10 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) */ if (newsize != oldsize) { inode_inc_iversion(inode); - if (!(mask & (ATTR_CTIME | ATTR_MTIME))) - inode->i_ctime = inode->i_mtime = - current_time(inode); + if (!(mask & (ATTR_CTIME | ATTR_MTIME))) { + inode->i_mtime = current_time(inode); + inode->i_ctime = inode->i_mtime; + } } if (newsize > oldsize) { @@ -4671,42 +5167,48 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) * truncation, it must capture all writes that happened before * this truncation. */ - btrfs_wait_for_snapshot_creation(root); - ret = btrfs_cont_expand(inode, oldsize, newsize); + btrfs_drew_write_lock(&root->snapshot_lock); + ret = btrfs_cont_expand(BTRFS_I(inode), oldsize, newsize); if (ret) { - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); return ret; } trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); return PTR_ERR(trans); } i_size_write(inode, newsize); - btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); + btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0); pagecache_isize_extended(inode, oldsize, newsize); - ret = btrfs_update_inode(trans, root, inode); - btrfs_end_write_no_snapshotting(root); + ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); + btrfs_drew_write_unlock(&root->snapshot_lock); btrfs_end_transaction(trans); } else { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + + if (btrfs_is_zoned(fs_info)) { + ret = btrfs_wait_ordered_range(inode, + ALIGN(newsize, fs_info->sectorsize), + (u64)-1); + if (ret) + return ret; + } /* * We're truncating a file that used to have good data down to - * zero. Make sure it gets into the ordered flush list so that - * any new writes get down to disk quickly. + * zero. Make sure any new writes to the file get on disk + * on close. */ if (newsize == 0) - set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE, + set_bit(BTRFS_INODE_FLUSH_ON_CLOSE, &BTRFS_I(inode)->runtime_flags); truncate_setsize(inode, newsize); - /* Disable nonlocked read DIO to avoid the endless truncate */ - btrfs_inode_block_unlocked_dio(BTRFS_I(inode)); inode_dio_wait(inode); - btrfs_inode_resume_unlocked_dio(BTRFS_I(inode)); ret = btrfs_truncate(inode, newsize == oldsize); if (ret && inode->i_nlink) { @@ -4728,7 +5230,8 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) return ret; } -static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) +static int btrfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct btrfs_root *root = BTRFS_I(inode)->root; @@ -4737,7 +5240,7 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) if (btrfs_root_readonly(root)) return -EROFS; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(mnt_userns, dentry, attr); if (err) return err; @@ -4748,60 +5251,44 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) } if (attr->ia_valid) { - setattr_copy(inode, attr); + setattr_copy(mnt_userns, inode, attr); inode_inc_iversion(inode); err = btrfs_dirty_inode(inode); if (!err && attr->ia_valid & ATTR_MODE) - err = posix_acl_chmod(inode, inode->i_mode); + err = posix_acl_chmod(mnt_userns, inode, inode->i_mode); } return err; } /* - * While truncating the inode pages during eviction, we get the VFS calling - * btrfs_invalidatepage() against each page of the inode. This is slow because - * the calls to btrfs_invalidatepage() result in a huge amount of calls to - * lock_extent_bits() and clear_extent_bit(), which keep merging and splitting - * extent_state structures over and over, wasting lots of time. + * While truncating the inode pages during eviction, we get the VFS + * calling btrfs_invalidate_folio() against each folio of the inode. This + * is slow because the calls to btrfs_invalidate_folio() result in a + * huge amount of calls to lock_extent() and clear_extent_bit(), + * which keep merging and splitting extent_state structures over and over, + * wasting lots of time. * - * Therefore if the inode is being evicted, let btrfs_invalidatepage() skip all - * those expensive operations on a per page basis and do only the ordered io - * finishing, while we release here the extent_map and extent_state structures, - * without the excessive merging and splitting. + * Therefore if the inode is being evicted, let btrfs_invalidate_folio() + * skip all those expensive operations on a per folio basis and do only + * the ordered io finishing, while we release here the extent_map and + * extent_state structures, without the excessive merging and splitting. */ static void evict_inode_truncate_pages(struct inode *inode) { struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - struct extent_map_tree *map_tree = &BTRFS_I(inode)->extent_tree; struct rb_node *node; ASSERT(inode->i_state & I_FREEING); truncate_inode_pages_final(&inode->i_data); - write_lock(&map_tree->lock); - while (!RB_EMPTY_ROOT(&map_tree->map.rb_root)) { - struct extent_map *em; - - node = rb_first_cached(&map_tree->map); - em = rb_entry(node, struct extent_map, rb_node); - clear_bit(EXTENT_FLAG_PINNED, &em->flags); - clear_bit(EXTENT_FLAG_LOGGING, &em->flags); - remove_extent_mapping(map_tree, em); - free_extent_map(em); - if (need_resched()) { - write_unlock(&map_tree->lock); - cond_resched(); - write_lock(&map_tree->lock); - } - } - write_unlock(&map_tree->lock); + btrfs_drop_extent_map_range(BTRFS_I(inode), 0, (u64)-1, false); /* * Keep looping until we have no more ranges in the io tree. - * We can have ongoing bios started by readpages (called from readahead) - * that have their endio callback (extent_io.c:end_bio_extent_readpage) + * We can have ongoing bios started by readahead that have + * their endio callback (extent_io.c:end_bio_extent_readpage) * still in progress (unlocked the pages in the bio but did not yet * unlocked the ranges in the io tree). Therefore this means some * ranges can still be locked and eviction started because before @@ -4829,22 +5316,22 @@ static void evict_inode_truncate_pages(struct inode *inode) state_flags = state->state; spin_unlock(&io_tree->lock); - lock_extent_bits(io_tree, start, end, &cached_state); + lock_extent(io_tree, start, end, &cached_state); /* * If still has DELALLOC flag, the extent didn't reach disk, * and its reserved space won't be freed by delayed_ref. * So we need to free its reserved space here. - * (Refer to comment in btrfs_invalidatepage, case 2) + * (Refer to comment in btrfs_invalidate_folio, case 2) * * Note, end is the bytenr of last byte, so we need + 1 here. */ if (state_flags & EXTENT_DELALLOC) - btrfs_qgroup_free_data(inode, NULL, start, end - start + 1); + btrfs_qgroup_free_data(BTRFS_I(inode), NULL, start, + end - start + 1); clear_extent_bit(io_tree, start, end, - EXTENT_LOCKED | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 1, 1, + EXTENT_CLEAR_ALL_BITS | EXTENT_DO_ACCOUNTING, &cached_state); cond_resched(); @@ -4857,7 +5344,6 @@ static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root, struct btrfs_block_rsv *rsv) { struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; struct btrfs_trans_handle *trans; u64 delayed_refs_extra = btrfs_calc_insert_metadata_size(fs_info, 1); int ret; @@ -4872,18 +5358,16 @@ static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root, * above. We reserve our extra bit here because we generate a ton of * delayed refs activity by truncating. * - * If we cannot make our reservation we'll attempt to steal from the - * global reserve, because we really want to be able to free up space. + * BTRFS_RESERVE_FLUSH_EVICT will steal from the global_rsv if it can, + * if we fail to make this reservation we can re-try without the + * delayed_refs_extra so we can make some forward progress. */ - ret = btrfs_block_rsv_refill(root, rsv, rsv->size + delayed_refs_extra, + ret = btrfs_block_rsv_refill(fs_info, rsv, rsv->size + delayed_refs_extra, BTRFS_RESERVE_FLUSH_EVICT); if (ret) { - /* - * Try to steal from the global reserve if there is space for - * it. - */ - if (btrfs_check_space_for_delayed_refs(fs_info) || - btrfs_block_rsv_migrate(global_rsv, rsv, rsv->size, 0)) { + ret = btrfs_block_rsv_refill(fs_info, rsv, rsv->size, + BTRFS_RESERVE_FLUSH_EVICT); + if (ret) { btrfs_warn(fs_info, "could not allocate space for delete; will truncate on mount"); return ERR_PTR(-ENOSPC); @@ -4915,6 +5399,7 @@ void btrfs_evict_inode(struct inode *inode) trace_btrfs_inode_evict(inode); if (!root) { + fsverity_cleanup_inode(inode); clear_inode(inode); return; } @@ -4941,26 +5426,45 @@ void btrfs_evict_inode(struct inode *inode) goto no_delete; } + /* + * This makes sure the inode item in tree is uptodate and the space for + * the inode update is released. + */ ret = btrfs_commit_inode_delayed_inode(BTRFS_I(inode)); if (ret) goto no_delete; + /* + * This drops any pending insert or delete operations we have for this + * inode. We could have a delayed dir index deletion queued up, but + * we're removing the inode completely so that'll be taken care of in + * the truncate. + */ + btrfs_kill_delayed_inode_items(BTRFS_I(inode)); + rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); if (!rsv) goto no_delete; rsv->size = btrfs_calc_metadata_size(fs_info, 1); - rsv->failfast = 1; + rsv->failfast = true; btrfs_i_size_write(BTRFS_I(inode), 0); while (1) { + struct btrfs_truncate_control control = { + .inode = BTRFS_I(inode), + .ino = btrfs_ino(BTRFS_I(inode)), + .new_size = 0, + .min_type = 0, + }; + trans = evict_refill_and_join(root, rsv); if (IS_ERR(trans)) goto free_rsv; trans->block_rsv = rsv; - ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0); + ret = btrfs_truncate_inode_items(trans, root, &control); trans->block_rsv = &fs_info->trans_block_rsv; btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); @@ -4987,10 +5491,6 @@ void btrfs_evict_inode(struct inode *inode) btrfs_end_transaction(trans); } - if (!(root == fs_info->tree_root || - root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)) - btrfs_return_ino(root, btrfs_ino(BTRFS_I(inode))); - free_rsv: btrfs_free_block_rsv(fs_info, rsv); no_delete: @@ -5000,6 +5500,7 @@ no_delete: * to retry these periodically in the future. */ btrfs_remove_delayed_node(BTRFS_I(inode)); + fsverity_cleanup_inode(inode); clear_inode(inode); } @@ -5098,7 +5599,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info, btrfs_release_path(path); - new_root = btrfs_read_fs_root_no_name(fs_info, location); + new_root = btrfs_get_fs_root(fs_info, location->objectid, true); if (IS_ERR(new_root)) { err = PTR_ERR(new_root); goto out; @@ -5150,15 +5651,15 @@ static void inode_tree_add(struct inode *inode) spin_unlock(&root->inode_lock); } -static void inode_tree_del(struct inode *inode) +static void inode_tree_del(struct btrfs_inode *inode) { - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; int empty = 0; spin_lock(&root->inode_lock); - if (!RB_EMPTY_NODE(&BTRFS_I(inode)->rb_node)) { - rb_erase(&BTRFS_I(inode)->rb_node, &root->inode_tree); - RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node); + if (!RB_EMPTY_NODE(&inode->rb_node)) { + rb_erase(&inode->rb_node, &root->inode_tree); + RB_CLEAR_NODE(&inode->rb_node); empty = RB_EMPTY_ROOT(&root->inode_tree); } spin_unlock(&root->inode_lock); @@ -5176,29 +5677,37 @@ static void inode_tree_del(struct inode *inode) static int btrfs_init_locked_inode(struct inode *inode, void *p) { struct btrfs_iget_args *args = p; - inode->i_ino = args->location->objectid; - memcpy(&BTRFS_I(inode)->location, args->location, - sizeof(*args->location)); - BTRFS_I(inode)->root = args->root; + + inode->i_ino = args->ino; + BTRFS_I(inode)->location.objectid = args->ino; + BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY; + BTRFS_I(inode)->location.offset = 0; + BTRFS_I(inode)->root = btrfs_grab_root(args->root); + BUG_ON(args->root && !BTRFS_I(inode)->root); + + if (args->root && args->root == args->root->fs_info->tree_root && + args->ino != BTRFS_BTREE_INODE_OBJECTID) + set_bit(BTRFS_INODE_FREE_SPACE_INODE, + &BTRFS_I(inode)->runtime_flags); return 0; } static int btrfs_find_actor(struct inode *inode, void *opaque) { struct btrfs_iget_args *args = opaque; - return args->location->objectid == BTRFS_I(inode)->location.objectid && + + return args->ino == BTRFS_I(inode)->location.objectid && args->root == BTRFS_I(inode)->root; } -static struct inode *btrfs_iget_locked(struct super_block *s, - struct btrfs_key *location, +static struct inode *btrfs_iget_locked(struct super_block *s, u64 ino, struct btrfs_root *root) { struct inode *inode; struct btrfs_iget_args args; - unsigned long hashval = btrfs_inode_hash(location->objectid, root); + unsigned long hashval = btrfs_inode_hash(ino, root); - args.location = location; + args.ino = ino; args.root = root; inode = iget5_locked(s, hashval, btrfs_find_actor, @@ -5208,17 +5717,17 @@ static struct inode *btrfs_iget_locked(struct super_block *s, } /* - * Get an inode object given its location and corresponding root. + * Get an inode object given its inode number and corresponding root. * Path can be preallocated to prevent recursing back to iget through * allocator. NULL is also valid but may require an additional allocation * later. */ -struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location, +struct inode *btrfs_iget_path(struct super_block *s, u64 ino, struct btrfs_root *root, struct btrfs_path *path) { struct inode *inode; - inode = btrfs_iget_locked(s, location, root); + inode = btrfs_iget_locked(s, ino, root); if (!inode) return ERR_PTR(-ENOMEM); @@ -5245,10 +5754,9 @@ struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location, return inode; } -struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, - struct btrfs_root *root) +struct inode *btrfs_iget(struct super_block *s, u64 ino, struct btrfs_root *root) { - return btrfs_iget_path(s, location, root, NULL); + return btrfs_iget_path(s, ino, root, NULL); } static struct inode *new_simple_dir(struct super_block *s, @@ -5260,7 +5768,7 @@ static struct inode *new_simple_dir(struct super_block *s, if (!inode) return ERR_PTR(-ENOMEM); - BTRFS_I(inode)->root = root; + BTRFS_I(inode)->root = btrfs_grab_root(root); memcpy(&BTRFS_I(inode)->location, key, sizeof(*key)); set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags); @@ -5281,21 +5789,17 @@ static struct inode *new_simple_dir(struct super_block *s, return inode; } +static_assert(BTRFS_FT_UNKNOWN == FT_UNKNOWN); +static_assert(BTRFS_FT_REG_FILE == FT_REG_FILE); +static_assert(BTRFS_FT_DIR == FT_DIR); +static_assert(BTRFS_FT_CHRDEV == FT_CHRDEV); +static_assert(BTRFS_FT_BLKDEV == FT_BLKDEV); +static_assert(BTRFS_FT_FIFO == FT_FIFO); +static_assert(BTRFS_FT_SOCK == FT_SOCK); +static_assert(BTRFS_FT_SYMLINK == FT_SYMLINK); + static inline u8 btrfs_inode_type(struct inode *inode) { - /* - * Compile-time asserts that generic FT_* types still match - * BTRFS_FT_* types - */ - BUILD_BUG_ON(BTRFS_FT_UNKNOWN != FT_UNKNOWN); - BUILD_BUG_ON(BTRFS_FT_REG_FILE != FT_REG_FILE); - BUILD_BUG_ON(BTRFS_FT_DIR != FT_DIR); - BUILD_BUG_ON(BTRFS_FT_CHRDEV != FT_CHRDEV); - BUILD_BUG_ON(BTRFS_FT_BLKDEV != FT_BLKDEV); - BUILD_BUG_ON(BTRFS_FT_FIFO != FT_FIFO); - BUILD_BUG_ON(BTRFS_FT_SOCK != FT_SOCK); - BUILD_BUG_ON(BTRFS_FT_SYMLINK != FT_SYMLINK); - return fs_umode_to_ftype(inode->i_mode); } @@ -5307,7 +5811,6 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) struct btrfs_root *sub_root = root; struct btrfs_key location; u8 di_type = 0; - int index; int ret = 0; if (dentry->d_name.len > BTRFS_NAME_LEN) @@ -5318,7 +5821,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) return ERR_PTR(ret); if (location.type == BTRFS_INODE_ITEM_KEY) { - inode = btrfs_iget(dir->i_sb, &location, root); + inode = btrfs_iget(dir->i_sb, location.objectid, root); if (IS_ERR(inode)) return inode; @@ -5334,20 +5837,20 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) return inode; } - 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) inode = ERR_PTR(ret); else - inode = new_simple_dir(dir->i_sb, &location, sub_root); + inode = new_simple_dir(dir->i_sb, &location, root); } else { - inode = btrfs_iget(dir->i_sb, &location, sub_root); - } - srcu_read_unlock(&fs_info->subvol_srcu, index); + inode = btrfs_iget(dir->i_sb, location.objectid, sub_root); + btrfs_put_root(sub_root); + + if (IS_ERR(inode)) + return inode; - if (!IS_ERR(inode) && root != sub_root) { down_read(&fs_info->cleanup_work_sem); if (!sb_rdonly(inode->i_sb)) ret = btrfs_orphan_cleanup(sub_root); @@ -5453,8 +5956,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) struct list_head ins_list; struct list_head del_list; int ret; - struct extent_buffer *leaf; - int slot; char *name_ptr; int name_len; int entries = 0; @@ -5481,35 +5982,19 @@ again: key.offset = ctx->pos; key.objectid = btrfs_ino(BTRFS_I(inode)); - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); - if (ret < 0) - goto err; - - while (1) { + btrfs_for_each_slot(root, &key, &found_key, path, ret) { struct dir_entry *entry; - - leaf = path->nodes[0]; - slot = path->slots[0]; - if (slot >= btrfs_header_nritems(leaf)) { - ret = btrfs_next_leaf(root, path); - if (ret < 0) - goto err; - else if (ret > 0) - break; - continue; - } - - btrfs_item_key_to_cpu(leaf, &found_key, slot); + struct extent_buffer *leaf = path->nodes[0]; if (found_key.objectid != key.objectid) break; if (found_key.type != BTRFS_DIR_INDEX_KEY) break; if (found_key.offset < ctx->pos) - goto next; + continue; if (btrfs_should_delete_dir_index(&del_list, found_key.offset)) - goto next; - di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item); + continue; + di = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dir_item); name_len = btrfs_dir_name_len(leaf, di); if ((total_len + sizeof(struct dir_entry) + name_len) >= PAGE_SIZE) { @@ -5536,9 +6021,11 @@ again: entries++; addr += sizeof(struct dir_entry) + name_len; total_len += sizeof(struct dir_entry) + name_len; -next: - path->slots[0]++; } + /* Catch error encountered during iteration */ + if (ret < 0) + goto err; + btrfs_release_path(path); ret = btrfs_filldir(private->filldir_buf, entries, ctx); @@ -5599,15 +6086,15 @@ static int btrfs_dirty_inode(struct inode *inode) if (IS_ERR(trans)) return PTR_ERR(trans); - ret = btrfs_update_inode(trans, root, inode); - if (ret && ret == -ENOSPC) { + ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); + if (ret && (ret == -ENOSPC || ret == -EDQUOT)) { /* whoops, lets try again with the full transaction */ 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); + ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); } btrfs_end_transaction(trans); if (BTRFS_I(inode)->delayed_node) @@ -5669,14 +6156,8 @@ static int btrfs_set_inode_index_count(struct btrfs_inode *inode) goto out; ret = 0; - /* - * MAGIC NUMBER EXPLANATION: - * since we search a directory based on f_pos we have to start at 2 - * since '.' and '..' have f_pos of 0 and 1 respectively, so everybody - * else has to start at 2 - */ if (path->slots[0] == 0) { - inode->index_cnt = 2; + inode->index_cnt = BTRFS_DIR_START_INDEX; goto out; } @@ -5687,7 +6168,7 @@ static int btrfs_set_inode_index_count(struct btrfs_inode *inode) if (found_key.objectid != btrfs_ino(inode) || found_key.type != BTRFS_DIR_INDEX_KEY) { - inode->index_cnt = 2; + inode->index_cnt = BTRFS_DIR_START_INDEX; goto out; } @@ -5723,7 +6204,8 @@ int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index) static int btrfs_insert_inode_locked(struct inode *inode) { struct btrfs_iget_args args; - args.location = &BTRFS_I(inode)->location; + + args.ino = BTRFS_I(inode)->location.objectid; args.root = BTRFS_I(inode)->root; return insert_inode_locked4(inode, @@ -5731,6 +6213,57 @@ static int btrfs_insert_inode_locked(struct inode *inode) btrfs_find_actor, &args); } +int btrfs_new_inode_prepare(struct btrfs_new_inode_args *args, + unsigned int *trans_num_items) +{ + struct inode *dir = args->dir; + struct inode *inode = args->inode; + int ret; + + ret = posix_acl_create(dir, &inode->i_mode, &args->default_acl, &args->acl); + if (ret) + return ret; + + /* 1 to add inode item */ + *trans_num_items = 1; + /* 1 to add compression property */ + if (BTRFS_I(dir)->prop_compress) + (*trans_num_items)++; + /* 1 to add default ACL xattr */ + if (args->default_acl) + (*trans_num_items)++; + /* 1 to add access ACL xattr */ + if (args->acl) + (*trans_num_items)++; +#ifdef CONFIG_SECURITY + /* 1 to add LSM xattr */ + if (dir->i_security) + (*trans_num_items)++; +#endif + if (args->orphan) { + /* 1 to add orphan item */ + (*trans_num_items)++; + } else { + /* + * 1 to add dir item + * 1 to add dir index + * 1 to update parent inode item + * + * No need for 1 unit for the inode ref item because it is + * inserted in a batch together with the inode item at + * btrfs_create_new_inode(). + */ + *trans_num_items += 3; + } + return 0; +} + +void btrfs_new_inode_args_destroy(struct btrfs_new_inode_args *args) +{ + posix_acl_release(args->acl); + posix_acl_release(args->default_acl); +} + /* * Inherit flags from the parent inode. * @@ -5740,9 +6273,6 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir) { unsigned int flags; - if (!dir) - return; - flags = BTRFS_I(dir)->flags; if (flags & BTRFS_INODE_NOCOMPRESS) { @@ -5762,81 +6292,92 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir) btrfs_sync_inode_flags_to_i_flags(inode); } -static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct inode *dir, - const char *name, int name_len, - u64 ref_objectid, u64 objectid, - umode_t mode, u64 *index) +int btrfs_create_new_inode(struct btrfs_trans_handle *trans, + struct btrfs_new_inode_args *args) { - struct btrfs_fs_info *fs_info = root->fs_info; - struct inode *inode; + struct inode *dir = args->dir; + struct inode *inode = args->inode; + const char *name = args->orphan ? NULL : args->dentry->d_name.name; + int name_len = args->orphan ? 0 : args->dentry->d_name.len; + struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); + struct btrfs_root *root; struct btrfs_inode_item *inode_item; struct btrfs_key *location; struct btrfs_path *path; + u64 objectid; struct btrfs_inode_ref *ref; struct btrfs_key key[2]; u32 sizes[2]; - int nitems = name ? 2 : 1; + struct btrfs_item_batch batch; unsigned long ptr; - unsigned int nofs_flag; int ret; path = btrfs_alloc_path(); if (!path) - return ERR_PTR(-ENOMEM); - - nofs_flag = memalloc_nofs_save(); - inode = new_inode(fs_info->sb); - memalloc_nofs_restore(nofs_flag); - if (!inode) { - btrfs_free_path(path); - return ERR_PTR(-ENOMEM); - } + return -ENOMEM; - /* - * O_TMPFILE, set link count to 0, so that after this point, - * we fill in an inode item with the correct link count. - */ - if (!name) - set_nlink(inode, 0); + if (!args->subvol) + BTRFS_I(inode)->root = btrfs_grab_root(BTRFS_I(dir)->root); + root = BTRFS_I(inode)->root; - /* - * we have to initialize this early, so we can reclaim the inode - * number if we fail afterwards in this function. - */ + ret = btrfs_get_free_objectid(root, &objectid); + if (ret) + goto out; inode->i_ino = objectid; - if (dir && name) { + if (args->orphan) { + /* + * O_TMPFILE, set link count to 0, so that after this point, we + * fill in an inode item with the correct link count. + */ + set_nlink(inode, 0); + } else { trace_btrfs_inode_request(dir); - ret = btrfs_set_inode_index(BTRFS_I(dir), index); - if (ret) { - btrfs_free_path(path); - iput(inode); - return ERR_PTR(ret); - } - } else if (dir) { - *index = 0; + ret = btrfs_set_inode_index(BTRFS_I(dir), &BTRFS_I(inode)->dir_index); + if (ret) + goto out; } - /* - * index_cnt is ignored for everything but a dir, - * btrfs_set_inode_index_count has an explanation for the magic - * number - */ - BTRFS_I(inode)->index_cnt = 2; - BTRFS_I(inode)->dir_index = *index; - BTRFS_I(inode)->root = root; + /* index_cnt is ignored for everything but a dir. */ + BTRFS_I(inode)->index_cnt = BTRFS_DIR_START_INDEX; BTRFS_I(inode)->generation = trans->transid; inode->i_generation = BTRFS_I(inode)->generation; /* + * Subvolumes don't inherit flags from their parent directory. + * Originally this was probably by accident, but we probably can't + * change it now without compatibility issues. + */ + if (!args->subvol) + btrfs_inherit_iflags(inode, dir); + + if (S_ISREG(inode->i_mode)) { + if (btrfs_test_opt(fs_info, NODATASUM)) + BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM; + if (btrfs_test_opt(fs_info, NODATACOW)) + BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW | + BTRFS_INODE_NODATASUM; + } + + location = &BTRFS_I(inode)->location; + location->objectid = objectid; + location->offset = 0; + location->type = BTRFS_INODE_ITEM_KEY; + + ret = btrfs_insert_inode_locked(inode); + if (ret < 0) { + if (!args->orphan) + BTRFS_I(dir)->index_cnt--; + goto out; + } + + /* * We could have gotten an inode number from somebody who was fsynced * and then removed in this same transaction, so let's just set full * sync since it will be a full sync anyway and this will blow away the * old info in the log. */ - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); + btrfs_set_inode_full_sync(BTRFS_I(inode)); key[0].objectid = objectid; key[0].type = BTRFS_INODE_ITEM_KEY; @@ -5844,7 +6385,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, sizes[0] = sizeof(struct btrfs_inode_item); - if (name) { + if (!args->orphan) { /* * Start new inodes with an inode_ref. This is slightly more * efficient for small numbers of hard links since they will @@ -5853,85 +6394,133 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, */ key[1].objectid = objectid; key[1].type = BTRFS_INODE_REF_KEY; - key[1].offset = ref_objectid; - - sizes[1] = name_len + sizeof(*ref); + if (args->subvol) { + key[1].offset = objectid; + sizes[1] = 2 + sizeof(*ref); + } else { + key[1].offset = btrfs_ino(BTRFS_I(dir)); + sizes[1] = name_len + sizeof(*ref); + } } - location = &BTRFS_I(inode)->location; - location->objectid = objectid; - location->offset = 0; - location->type = BTRFS_INODE_ITEM_KEY; - - ret = btrfs_insert_inode_locked(inode); - if (ret < 0) { - iput(inode); - goto fail; + batch.keys = &key[0]; + batch.data_sizes = &sizes[0]; + batch.total_data_size = sizes[0] + (args->orphan ? 0 : sizes[1]); + batch.nr = args->orphan ? 1 : 2; + ret = btrfs_insert_empty_items(trans, root, path, &batch); + if (ret != 0) { + btrfs_abort_transaction(trans, ret); + goto discard; } - path->leave_spinning = 1; - ret = btrfs_insert_empty_items(trans, root, path, key, sizes, nitems); - if (ret != 0) - goto fail_unlock; - - inode_init_owner(inode, dir, mode); - inode_set_bytes(inode, 0); - inode->i_mtime = current_time(inode); inode->i_atime = inode->i_mtime; inode->i_ctime = inode->i_mtime; BTRFS_I(inode)->i_otime = inode->i_mtime; + /* + * We're going to fill the inode item now, so at this point the inode + * must be fully initialized. + */ + inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_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); - if (name) { + if (!args->orphan) { ref = btrfs_item_ptr(path->nodes[0], path->slots[0] + 1, struct btrfs_inode_ref); - btrfs_set_inode_ref_name_len(path->nodes[0], ref, name_len); - btrfs_set_inode_ref_index(path->nodes[0], ref, *index); ptr = (unsigned long)(ref + 1); - write_extent_buffer(path->nodes[0], name, ptr, name_len); + if (args->subvol) { + btrfs_set_inode_ref_name_len(path->nodes[0], ref, 2); + btrfs_set_inode_ref_index(path->nodes[0], ref, 0); + write_extent_buffer(path->nodes[0], "..", ptr, 2); + } else { + btrfs_set_inode_ref_name_len(path->nodes[0], ref, name_len); + btrfs_set_inode_ref_index(path->nodes[0], ref, + BTRFS_I(inode)->dir_index); + write_extent_buffer(path->nodes[0], name, ptr, name_len); + } } btrfs_mark_buffer_dirty(path->nodes[0]); + /* + * We don't need the path anymore, plus inheriting properties, adding + * ACLs, security xattrs, orphan item or adding the link, will result in + * allocating yet another path. So just free our path. + */ btrfs_free_path(path); + path = NULL; - btrfs_inherit_iflags(inode, dir); + if (args->subvol) { + struct inode *parent; - if (S_ISREG(mode)) { - if (btrfs_test_opt(fs_info, NODATASUM)) - BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM; - if (btrfs_test_opt(fs_info, NODATACOW)) - BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW | - BTRFS_INODE_NODATASUM; + /* + * Subvolumes inherit properties from their parent subvolume, + * not the directory they were created in. + */ + parent = btrfs_iget(fs_info->sb, BTRFS_FIRST_FREE_OBJECTID, + BTRFS_I(dir)->root); + if (IS_ERR(parent)) { + ret = PTR_ERR(parent); + } else { + ret = btrfs_inode_inherit_props(trans, inode, parent); + iput(parent); + } + } else { + ret = btrfs_inode_inherit_props(trans, inode, dir); + } + if (ret) { + btrfs_err(fs_info, + "error inheriting props for ino %llu (root %llu): %d", + btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, + ret); + } + + /* + * Subvolumes don't inherit ACLs or get passed to the LSM. This is + * probably a bug. + */ + if (!args->subvol) { + ret = btrfs_init_inode_security(trans, args); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto discard; + } } inode_tree_add(inode); trace_btrfs_inode_new(inode); - btrfs_set_inode_last_trans(trans, inode); + btrfs_set_inode_last_trans(trans, BTRFS_I(inode)); btrfs_update_root_times(trans, root); - ret = btrfs_inode_inherit_props(trans, inode, dir); - if (ret) - btrfs_err(fs_info, - "error inheriting props for ino %llu (root %llu): %d", - btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, ret); + if (args->orphan) { + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); + } else { + ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), name, + name_len, 0, BTRFS_I(inode)->dir_index); + } + if (ret) { + btrfs_abort_transaction(trans, ret); + goto discard; + } - return inode; + return 0; -fail_unlock: +discard: + /* + * discard_new_inode() calls iput(), but the caller owns the reference + * to the inode. + */ + ihold(inode); discard_new_inode(inode); -fail: - if (dir && name) - BTRFS_I(dir)->index_cnt--; +out: btrfs_free_path(path); - return ERR_PTR(ret); + return ret; } /* @@ -5995,7 +6584,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, parent_inode->vfs_inode.i_mtime = now; parent_inode->vfs_inode.i_ctime = now; } - ret = btrfs_update_inode(trans, root, &parent_inode->vfs_inode); + ret = btrfs_update_inode(trans, root, parent_inode); if (ret) btrfs_abort_transaction(trans, ret); return ret; @@ -6023,148 +6612,71 @@ fail_dir_item: return ret; } -static int btrfs_add_nondir(struct btrfs_trans_handle *trans, - 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, - backref, index); - if (err > 0) - err = -EEXIST; - return err; -} - -static int btrfs_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int btrfs_create_common(struct inode *dir, struct dentry *dentry, + struct inode *inode) { 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; + struct btrfs_new_inode_args new_inode_args = { + .dir = dir, + .dentry = dentry, + .inode = inode, + }; + unsigned int trans_num_items; + struct btrfs_trans_handle *trans; int err; - u64 objectid; - u64 index = 0; - - /* - * 2 for inode item and ref - * 2 for dir items - * 1 for xattr if selinux is on - */ - trans = btrfs_start_transaction(root, 5); - if (IS_ERR(trans)) - return PTR_ERR(trans); - err = btrfs_find_free_ino(root, &objectid); + err = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items); if (err) - goto out_unlock; + goto out_inode; - inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, - mode, &index); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - inode = NULL; - goto out_unlock; + trans = btrfs_start_transaction(root, trans_num_items); + if (IS_ERR(trans)) { + err = PTR_ERR(trans); + goto out_new_inode_args; } - /* - * If the active LSM wants to access the inode during - * d_instantiate it needs these. Smack checks to see - * if the filesystem supports xattrs by looking at the - * ops vector. - */ - inode->i_op = &btrfs_special_inode_operations; - init_special_inode(inode, inode->i_mode, rdev); - - err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); - if (err) - goto out_unlock; - - err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode), - 0, index); - if (err) - goto out_unlock; - - btrfs_update_inode(trans, root, inode); - d_instantiate_new(dentry, inode); + err = btrfs_create_new_inode(trans, &new_inode_args); + if (!err) + d_instantiate_new(dentry, inode); -out_unlock: btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); - if (err && inode) { - inode_dec_link_count(inode); - discard_new_inode(inode); - } +out_new_inode_args: + btrfs_new_inode_args_destroy(&new_inode_args); +out_inode: + if (err) + iput(inode); return err; } -static int btrfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int btrfs_mknod(struct user_namespace *mnt_userns, 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; - int err; - u64 objectid; - u64 index = 0; + struct inode *inode; - /* - * 2 for inode item and ref - * 2 for dir items - * 1 for xattr if selinux is on - */ - trans = btrfs_start_transaction(root, 5); - if (IS_ERR(trans)) - return PTR_ERR(trans); + inode = new_inode(dir->i_sb); + if (!inode) + return -ENOMEM; + inode_init_owner(mnt_userns, inode, dir, mode); + inode->i_op = &btrfs_special_inode_operations; + init_special_inode(inode, inode->i_mode, rdev); + return btrfs_create_common(dir, dentry, inode); +} - err = btrfs_find_free_ino(root, &objectid); - if (err) - goto out_unlock; +static int btrfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) +{ + struct inode *inode; - inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, - mode, &index); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - inode = NULL; - goto out_unlock; - } - /* - * If the active LSM wants to access the inode during - * d_instantiate it needs these. Smack checks to see - * if the filesystem supports xattrs by looking at the - * ops vector. - */ + inode = new_inode(dir->i_sb); + if (!inode) + return -ENOMEM; + inode_init_owner(mnt_userns, inode, dir, mode); inode->i_fop = &btrfs_file_operations; inode->i_op = &btrfs_file_inode_operations; inode->i_mapping->a_ops = &btrfs_aops; - - err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); - if (err) - goto out_unlock; - - err = btrfs_update_inode(trans, root, inode); - if (err) - goto out_unlock; - - err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode), - 0, index); - if (err) - goto out_unlock; - - BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - d_instantiate_new(dentry, inode); - -out_unlock: - btrfs_end_transaction(trans); - if (err && inode) { - inode_dec_link_count(inode); - discard_new_inode(inode); - } - btrfs_btree_balance_dirty(fs_info); - return err; + return btrfs_create_common(dir, dentry, inode); } static int btrfs_link(struct dentry *old_dentry, struct inode *dir, @@ -6210,16 +6722,15 @@ 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, BTRFS_I(dir), dentry, BTRFS_I(inode), - 1, index); + err = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), + dentry->d_name.name, dentry->d_name.len, 1, index); if (err) { drop_inode = 1; } else { struct dentry *parent = dentry->d_parent; - int ret; - err = btrfs_update_inode(trans, root, inode); + err = btrfs_update_inode(trans, root, BTRFS_I(inode)); if (err) goto fail; if (inode->i_nlink == 1) { @@ -6232,12 +6743,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, goto fail; } d_instantiate(dentry, inode); - ret = btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent, - true, NULL); - if (ret == BTRFS_NEED_TRANS_COMMIT) { - err = btrfs_commit_transaction(trans); - trans = NULL; - } + btrfs_log_new_name(trans, old_dentry, NULL, 0, parent); } fail: @@ -6251,67 +6757,18 @@ fail: return err; } -static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int btrfs_mkdir(struct user_namespace *mnt_userns, 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; - int err = 0; - u64 objectid = 0; - u64 index = 0; - - /* - * 2 items for inode and ref - * 2 items for dir items - * 1 for xattr if selinux is on - */ - trans = btrfs_start_transaction(root, 5); - if (IS_ERR(trans)) - return PTR_ERR(trans); - - err = btrfs_find_free_ino(root, &objectid); - if (err) - goto out_fail; - - inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid, - S_IFDIR | mode, &index); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - inode = NULL; - goto out_fail; - } + struct inode *inode; - /* these must be set before we unlock the inode */ + inode = new_inode(dir->i_sb); + if (!inode) + return -ENOMEM; + inode_init_owner(mnt_userns, inode, dir, S_IFDIR | mode); inode->i_op = &btrfs_dir_inode_operations; inode->i_fop = &btrfs_dir_file_operations; - - err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); - if (err) - goto out_fail; - - btrfs_i_size_write(BTRFS_I(inode), 0); - err = btrfs_update_inode(trans, root, inode); - if (err) - goto out_fail; - - 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; - - d_instantiate_new(dentry, inode); - -out_fail: - btrfs_end_transaction(trans); - if (err && inode) { - inode_dec_link_count(inode); - discard_new_inode(inode); - } - btrfs_btree_balance_dirty(fs_info); - return err; + return btrfs_create_common(dir, dentry, inode); } static noinline int uncompress_inline(struct btrfs_path *path, @@ -6330,8 +6787,7 @@ static noinline int uncompress_inline(struct btrfs_path *path, WARN_ON(pg_offset != 0); compress_type = btrfs_file_extent_compression(leaf, item); max_size = btrfs_file_extent_ram_bytes(leaf, item); - inline_size = btrfs_file_extent_inline_item_len(leaf, - btrfs_item_nr(path->slots[0])); + inline_size = btrfs_file_extent_inline_item_len(leaf, path->slots[0]); tmp = kmalloc(inline_size, GFP_NOFS); if (!tmp) return -ENOMEM; @@ -6351,11 +6807,9 @@ static noinline int uncompress_inline(struct btrfs_path *path, * 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); - } + if (max_size + pg_offset < PAGE_SIZE) + memzero_page(page, pg_offset + max_size, + PAGE_SIZE - max_size - pg_offset); kfree(tmp); return ret; } @@ -6383,8 +6837,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, u64 start, u64 len) { struct btrfs_fs_info *fs_info = inode->root->fs_info; - int ret; - int err = 0; + int ret = 0; u64 extent_start = 0; u64 extent_end = 0; u64 objectid = btrfs_ino(inode); @@ -6396,7 +6849,6 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, struct btrfs_key found_key; struct extent_map *em = NULL; struct extent_map_tree *em_tree = &inode->extent_tree; - struct extent_io_tree *io_tree = &inode->io_tree; read_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, start, len); @@ -6412,7 +6864,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, } em = alloc_extent_map(); if (!em) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } em->start = EXTENT_MAP_HOLE; @@ -6422,7 +6874,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, path = btrfs_alloc_path(); if (!path) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } @@ -6430,19 +6882,23 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, path->reada = READA_FORWARD; /* - * Unless we're going to uncompress the inline extent, no sleep would - * happen. + * The same explanation in load_free_space_cache applies here as well, + * we only read when we're loading the free space cache, and at that + * point the commit_root has everything we need. */ - path->leave_spinning = 1; + if (btrfs_is_free_space_inode(inode)) { + path->search_commit_root = 1; + path->skip_locking = 1; + } ret = btrfs_lookup_file_extent(NULL, root, path, objectid, start, 0); if (ret < 0) { - err = ret; goto out; } else if (ret > 0) { if (path->slots[0] == 0) goto not_found; path->slots[0]--; + ret = 0; } leaf = path->nodes[0]; @@ -6463,6 +6919,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, extent_type = btrfs_file_extent_type(leaf, item); extent_start = found_key.offset; + extent_end = btrfs_file_extent_end(path); if (extent_type == BTRFS_FILE_EXTENT_REG || extent_type == BTRFS_FILE_EXTENT_PREALLOC) { /* Only regular file could have regular/prealloc extent */ @@ -6473,18 +6930,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, btrfs_ino(inode)); goto out; } - extent_end = extent_start + - btrfs_file_extent_num_bytes(leaf, item); - trace_btrfs_get_extent_show_fi_regular(inode, leaf, item, extent_start); } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - size_t size; - - size = btrfs_file_extent_ram_bytes(leaf, item); - extent_end = ALIGN(extent_start + size, - fs_info->sectorsize); - trace_btrfs_get_extent_show_fi_inline(inode, leaf, item, path->slots[0], extent_start); @@ -6494,12 +6942,11 @@ next: path->slots[0]++; if (path->slots[0] >= btrfs_header_nritems(leaf)) { ret = btrfs_next_leaf(root, path); - if (ret < 0) { - err = ret; + if (ret < 0) goto out; - } else if (ret > 0) { + else if (ret > 0) goto not_found; - } + leaf = path->nodes[0]; } btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); @@ -6544,18 +6991,15 @@ next: em->orig_start = em->start; ptr = btrfs_file_extent_inline_start(item) + extent_offset; - btrfs_set_path_blocking(path); if (!PageUptodate(page)) { if (btrfs_file_extent_compression(leaf, item) != BTRFS_COMPRESS_NONE) { ret = uncompress_inline(path, page, pg_offset, extent_offset, item); - if (ret) { - err = ret; + if (ret) goto out; - } } else { - map = kmap(page); + map = kmap_local_page(page); read_extent_buffer(leaf, map + pg_offset, ptr, copy_size); if (pg_offset + copy_size < PAGE_SIZE) { @@ -6563,12 +7007,10 @@ next: PAGE_SIZE - pg_offset - copy_size); } - kunmap(page); + kunmap_local(map); } flush_dcache_page(page); } - set_extent_uptodate(io_tree, em->start, - extent_map_end(em) - 1, NULL, GFP_NOFS); goto insert; } not_found: @@ -6577,160 +7019,32 @@ not_found: em->len = len; em->block_start = EXTENT_MAP_HOLE; insert: + ret = 0; btrfs_release_path(path); if (em->start > start || extent_map_end(em) <= start) { btrfs_err(fs_info, "bad extent! em: [%llu %llu] passed [%llu %llu]", em->start, em->len, start, len); - err = -EIO; + ret = -EIO; goto out; } - err = 0; write_lock(&em_tree->lock); - err = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); write_unlock(&em_tree->lock); out: btrfs_free_path(path); trace_btrfs_get_extent(root, inode, em); - if (err) { - free_extent_map(em); - return ERR_PTR(err); - } - BUG_ON(!em); /* Error is always set */ - return em; -} - -struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode, - u64 start, u64 len) -{ - struct extent_map *em; - struct extent_map *hole_em = NULL; - u64 delalloc_start = start; - u64 end; - u64 delalloc_len; - u64 delalloc_end; - int err = 0; - - em = btrfs_get_extent(inode, NULL, 0, start, len); - if (IS_ERR(em)) - return em; - /* - * If our em maps to: - * - a hole or - * - a pre-alloc extent, - * there might actually be delalloc bytes behind it. - */ - if (em->block_start != EXTENT_MAP_HOLE && - !test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - return em; - else - hole_em = em; - - /* check to see if we've wrapped (len == -1 or similar) */ - end = start + len; - if (end < start) - end = (u64)-1; - else - end -= 1; - - em = NULL; - - /* ok, we didn't find anything, lets look for delalloc */ - delalloc_len = count_range_bits(&inode->io_tree, &delalloc_start, - end, len, EXTENT_DELALLOC, 1); - delalloc_end = delalloc_start + delalloc_len; - if (delalloc_end < delalloc_start) - delalloc_end = (u64)-1; - - /* - * We didn't find anything useful, return the original results from - * get_extent() - */ - if (delalloc_start > end || delalloc_end <= start) { - em = hole_em; - hole_em = NULL; - goto out; - } - - /* - * Adjust the delalloc_start to make sure it doesn't go backwards from - * the start they passed in - */ - delalloc_start = max(start, delalloc_start); - delalloc_len = delalloc_end - delalloc_start; - - if (delalloc_len > 0) { - u64 hole_start; - u64 hole_len; - const u64 hole_end = extent_map_end(hole_em); - - em = alloc_extent_map(); - if (!em) { - err = -ENOMEM; - goto out; - } - - ASSERT(hole_em); - /* - * When btrfs_get_extent can't find anything it returns one - * huge hole - * - * Make sure what it found really fits our range, and adjust to - * make sure it is based on the start from the caller - */ - if (hole_end <= start || hole_em->start > end) { - free_extent_map(hole_em); - hole_em = NULL; - } else { - hole_start = max(hole_em->start, start); - hole_len = hole_end - hole_start; - } - - if (hole_em && delalloc_start > hole_start) { - /* - * Our hole starts before our delalloc, so we have to - * return just the parts of the hole that go until the - * delalloc starts - */ - em->len = min(hole_len, delalloc_start - hole_start); - em->start = hole_start; - em->orig_start = hole_start; - /* - * Don't adjust block start at all, it is fixed at - * EXTENT_MAP_HOLE - */ - em->block_start = hole_em->block_start; - em->block_len = hole_len; - if (test_bit(EXTENT_FLAG_PREALLOC, &hole_em->flags)) - set_bit(EXTENT_FLAG_PREALLOC, &em->flags); - } else { - /* - * Hole is out of passed range or it starts after - * delalloc range - */ - em->start = delalloc_start; - em->len = delalloc_len; - em->orig_start = delalloc_start; - em->block_start = EXTENT_MAP_DELALLOC; - em->block_len = delalloc_len; - } - } else { - return hole_em; - } -out: - - free_extent_map(hole_em); - if (err) { + if (ret) { free_extent_map(em); - return ERR_PTR(err); + return ERR_PTR(ret); } return em; } -static struct extent_map *btrfs_create_dio_extent(struct inode *inode, +static struct extent_map *btrfs_create_dio_extent(struct btrfs_inode *inode, const u64 start, const u64 len, const u64 orig_start, @@ -6744,21 +7058,23 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, int ret; if (type != BTRFS_ORDERED_NOCOW) { - em = create_io_em(inode, start, len, orig_start, - block_start, block_len, orig_block_len, - ram_bytes, + 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; } - ret = btrfs_add_ordered_extent_dio(inode, start, block_start, - len, block_len, type); + ret = btrfs_add_ordered_extent(inode, start, len, len, block_start, + block_len, 0, + (1 << type) | + (1 << BTRFS_ORDERED_DIRECT), + BTRFS_COMPRESS_NONE); if (ret) { if (em) { free_extent_map(em); - btrfs_drop_extent_cache(BTRFS_I(inode), start, - start + len - 1, 0); + btrfs_drop_extent_map_range(inode, start, + start + len - 1, false); } em = ERR_PTR(ret); } @@ -6767,11 +7083,11 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, return em; } -static struct extent_map *btrfs_new_extent_direct(struct inode *inode, +static struct extent_map *btrfs_new_extent_direct(struct btrfs_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 btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct extent_map *em; struct btrfs_key ins; u64 alloc_hint; @@ -6788,21 +7104,51 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, ins.offset, BTRFS_ORDERED_REGULAR); btrfs_dec_block_group_reservations(fs_info, ins.objectid); if (IS_ERR(em)) - btrfs_free_reserved_extent(fs_info, ins.objectid, - ins.offset, 1); + btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, + 1); return em; } +static bool btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr) +{ + struct btrfs_block_group *block_group; + bool readonly = false; + + block_group = btrfs_lookup_block_group(fs_info, bytenr); + if (!block_group || block_group->ro) + readonly = true; + if (block_group) + btrfs_put_block_group(block_group); + return readonly; +} + /* - * returns 1 when the nocow is safe, < 1 on error, 0 if the - * block must be cow'd + * Check if we can do nocow write into the range [@offset, @offset + @len) + * + * @offset: File offset + * @len: The length to write, will be updated to the nocow writeable + * range + * @orig_start: (optional) Return the original file offset of the file extent + * @orig_len: (optional) Return the original on-disk length of the file extent + * @ram_bytes: (optional) Return the ram_bytes of the file extent + * @strict: if true, omit optimizations that might force us into unnecessary + * cow. e.g., don't trust generation number. + * + * Return: + * >0 and update @len if we can do nocow write + * 0 if we can't do nocow write + * <0 if error happened + * + * NOTE: This only checks the file extents, caller is responsible to wait for + * any ordered extents. */ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, u64 *orig_start, u64 *orig_block_len, - u64 *ram_bytes) + u64 *ram_bytes, bool nowait, bool strict) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct can_nocow_file_extent_args nocow_args = { 0 }; struct btrfs_path *path; int ret; struct extent_buffer *leaf; @@ -6810,35 +7156,29 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_file_extent_item *fi; struct btrfs_key key; - u64 disk_bytenr; - u64 backref_offset; - u64 extent_end; - u64 num_bytes; - int slot; int found_type; - bool nocow = (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW); path = btrfs_alloc_path(); if (!path) return -ENOMEM; + path->nowait = nowait; ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(BTRFS_I(inode)), offset, 0); if (ret < 0) goto out; - slot = path->slots[0]; if (ret == 1) { - if (slot == 0) { + if (path->slots[0] == 0) { /* can't find the item, must cow */ ret = 0; goto out; } - slot--; + path->slots[0]--; } ret = 0; leaf = path->nodes[0]; - btrfs_item_key_to_cpu(leaf, &key, slot); + btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); if (key.objectid != btrfs_ino(BTRFS_I(inode)) || key.type != BTRFS_EXTENT_DATA_KEY) { /* not our file or wrong item type, must cow */ @@ -6850,54 +7190,38 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, goto out; } - fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); - found_type = btrfs_file_extent_type(leaf, fi); - if (found_type != BTRFS_FILE_EXTENT_REG && - found_type != BTRFS_FILE_EXTENT_PREALLOC) { - /* not a regular extent, must cow */ - goto out; - } - - if (!nocow && found_type == BTRFS_FILE_EXTENT_REG) + if (btrfs_file_extent_end(path) <= offset) goto out; - extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); - if (extent_end <= offset) - goto out; + fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + found_type = btrfs_file_extent_type(leaf, fi); + if (ram_bytes) + *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); - disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); - if (disk_bytenr == 0) - goto out; + nocow_args.start = offset; + nocow_args.end = offset + *len - 1; + nocow_args.strict = strict; + nocow_args.free_path = true; - if (btrfs_file_extent_compression(leaf, fi) || - btrfs_file_extent_encryption(leaf, fi) || - btrfs_file_extent_other_encoding(leaf, fi)) - goto out; + ret = can_nocow_file_extent(path, &key, BTRFS_I(inode), &nocow_args); + /* can_nocow_file_extent() has freed the path. */ + path = NULL; - /* - * Do the same check as in btrfs_cross_ref_exist but without the - * unnecessary search. - */ - if (btrfs_file_extent_generation(leaf, fi) <= - btrfs_root_last_snapshot(&root->root_item)) + if (ret != 1) { + /* Treat errors as not being able to NOCOW. */ + ret = 0; goto out; - - backref_offset = btrfs_file_extent_offset(leaf, fi); - - if (orig_start) { - *orig_start = key.offset - backref_offset; - *orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi); - *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); } - if (btrfs_extent_readonly(fs_info, disk_bytenr)) + ret = 0; + if (btrfs_extent_readonly(fs_info, nocow_args.disk_bytenr)) goto out; - num_bytes = min(offset + *len, extent_end) - offset; - if (!nocow && found_type == BTRFS_FILE_EXTENT_PREALLOC) { + if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) && + found_type == BTRFS_FILE_EXTENT_PREALLOC) { u64 range_end; - range_end = round_up(offset + num_bytes, + range_end = round_up(offset + nocow_args.num_bytes, root->fs_info->sectorsize) - 1; ret = test_range_bit(io_tree, offset, range_end, EXTENT_DELALLOC, 0, NULL); @@ -6907,35 +7231,12 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, } } - btrfs_release_path(path); - - /* - * look for other files referencing this extent, if we - * find any we must cow - */ - - ret = btrfs_cross_ref_exist(root, btrfs_ino(BTRFS_I(inode)), - key.offset - backref_offset, disk_bytenr); - if (ret) { - ret = 0; - goto out; - } + if (orig_start) + *orig_start = key.offset - nocow_args.extent_offset; + if (orig_block_len) + *orig_block_len = nocow_args.disk_num_bytes; - /* - * adjust disk_bytenr and num_bytes to cover just the bytes - * in this extent we are about to write. If there - * are any csums in that range we have to cow in order - * to keep the csums correct - */ - disk_bytenr += backref_offset; - disk_bytenr += offset - key.offset; - 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 - */ - *len = num_bytes; + *len = nocow_args.num_bytes; ret = 1; out: btrfs_free_path(path); @@ -6943,14 +7244,22 @@ out: } static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, - struct extent_state **cached_state, int writing) + struct extent_state **cached_state, + unsigned int iomap_flags) { + const bool writing = (iomap_flags & IOMAP_WRITE); + const bool nowait = (iomap_flags & IOMAP_NOWAIT); + struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_ordered_extent *ordered; int ret = 0; while (1) { - lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend, - cached_state); + if (nowait) { + if (!try_lock_extent(io_tree, lockstart, lockend)) + return -EAGAIN; + } else { + lock_extent(io_tree, lockstart, lockend, cached_state); + } /* * We're concerned with the entire range that we're going to be * doing DIO to, so we need to make sure there's no ordered @@ -6971,10 +7280,14 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, lockstart, lockend))) break; - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, - cached_state); + unlock_extent(io_tree, lockstart, lockend, cached_state); if (ordered) { + if (nowait) { + btrfs_put_ordered_extent(ordered); + ret = -EAGAIN; + break; + } /* * If we are doing a DIO read and the ordered extent we * found is for a buffered write, we can not wait for it @@ -6992,9 +7305,9 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, */ if (writing || test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags)) - btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_start_ordered_extent(ordered, 1); else - ret = -ENOTBLK; + ret = nowait ? -EAGAIN : -ENOTBLK; btrfs_put_ordered_extent(ordered); } else { /* @@ -7002,15 +7315,15 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, * for it to complete) and then invalidate the pages for * this range (through invalidate_inode_pages2_range()), * but that can lead us to a deadlock with a concurrent - * call to readpages() (a buffered read or a defrag call + * call to readahead (a buffered read or a defrag call * triggered a readahead) on a page lock due to an * ordered dio extent we created before but did not have * yet a corresponding bio submitted (whence it can not - * complete), which makes readpages() wait for that + * complete), which makes readahead wait for that * ordered extent to complete while holding a lock on * that page. */ - ret = -ENOTBLK; + ret = nowait ? -EAGAIN : -ENOTBLK; } if (ret) @@ -7023,13 +7336,12 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, } /* 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, +static struct extent_map *create_io_em(struct btrfs_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; int ret; @@ -7038,7 +7350,6 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, type == BTRFS_ORDERED_NOCOW || type == BTRFS_ORDERED_REGULAR); - em_tree = &BTRFS_I(inode)->extent_tree; em = alloc_extent_map(); if (!em) return ERR_PTR(-ENOMEM); @@ -7059,18 +7370,7 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, em->compress_type = compress_type; } - do { - 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); - + ret = btrfs_replace_extent_map_range(inode, em, true); if (ret) { free_extent_map(em); return ERR_PTR(ret); @@ -7081,36 +7381,21 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, } -static int btrfs_get_blocks_direct_read(struct extent_map *em, - struct buffer_head *bh_result, - struct inode *inode, - u64 start, u64 len) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - - if (em->block_start == EXTENT_MAP_HOLE || - test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - return -ENOENT; - - len = min(len, em->len - (start - em->start)); - - bh_result->b_blocknr = (em->block_start + (start - em->start)) >> - inode->i_blkbits; - bh_result->b_size = len; - bh_result->b_bdev = fs_info->fs_devices->latest_bdev; - set_buffer_mapped(bh_result); - - return 0; -} - static int btrfs_get_blocks_direct_write(struct extent_map **map, - struct buffer_head *bh_result, struct inode *inode, struct btrfs_dio_data *dio_data, - u64 start, u64 len) + u64 start, u64 len, + unsigned int iomap_flags) { + const bool nowait = (iomap_flags & IOMAP_NOWAIT); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_map *em = *map; + int type; + u64 block_start, orig_start, orig_block_len, ram_bytes; + struct btrfs_block_group *bg; + bool can_nocow = false; + bool space_reserved = false; + u64 prev_len; int ret = 0; /* @@ -7125,9 +7410,6 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map, if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) || ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) && em->block_start != EXTENT_MAP_HOLE)) { - int type; - u64 block_start, orig_start, orig_block_len, ram_bytes; - if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) type = BTRFS_ORDERED_PREALLOC; else @@ -7136,108 +7418,207 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map, block_start = em->block_start + (start - em->start); if (can_nocow_extent(inode, start, &len, &orig_start, - &orig_block_len, &ram_bytes) == 1 && - 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(fs_info, block_start); - if (type == BTRFS_ORDERED_PREALLOC) { - free_extent_map(em); - *map = em = em2; - } - - if (em2 && IS_ERR(em2)) { - ret = PTR_ERR(em2); - goto out; - } - /* - * For inode marked NODATACOW or extent marked PREALLOC, - * use the existing or preallocated extent, so does not - * need to adjust btrfs_space_info's bytes_may_use. - */ - btrfs_free_reserved_data_space_noquota(inode, start, - len); - goto skip_cow; + &orig_block_len, &ram_bytes, false, false) == 1) { + bg = btrfs_inc_nocow_writers(fs_info, block_start); + if (bg) + can_nocow = true; } } - /* this will cow the extent */ - len = bh_result->b_size; - free_extent_map(em); - *map = em = btrfs_new_extent_direct(inode, start, len); - if (IS_ERR(em)) { - ret = PTR_ERR(em); - goto out; - } + prev_len = len; + if (can_nocow) { + struct extent_map *em2; - len = min(len, em->len - (start - em->start)); + /* We can NOCOW, so only need to reserve metadata space. */ + ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len, len, + nowait); + if (ret < 0) { + /* Our caller expects us to free the input extent map. */ + free_extent_map(em); + *map = NULL; + btrfs_dec_nocow_writers(bg); + if (nowait && (ret == -ENOSPC || ret == -EDQUOT)) + ret = -EAGAIN; + goto out; + } + space_reserved = true; -skip_cow: - bh_result->b_blocknr = (em->block_start + (start - em->start)) >> - inode->i_blkbits; - bh_result->b_size = len; - bh_result->b_bdev = fs_info->fs_devices->latest_bdev; - set_buffer_mapped(bh_result); + em2 = btrfs_create_dio_extent(BTRFS_I(inode), start, len, + orig_start, block_start, + len, orig_block_len, + ram_bytes, type); + btrfs_dec_nocow_writers(bg); + if (type == BTRFS_ORDERED_PREALLOC) { + free_extent_map(em); + *map = em2; + em = em2; + } - if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - set_buffer_new(bh_result); + if (IS_ERR(em2)) { + ret = PTR_ERR(em2); + goto out; + } + + dio_data->nocow_done = true; + } else { + /* Our caller expects us to free the input extent map. */ + free_extent_map(em); + *map = NULL; + + if (nowait) + return -EAGAIN; + + /* + * If we could not allocate data space before locking the file + * range and we can't do a NOCOW write, then we have to fail. + */ + if (!dio_data->data_space_reserved) + return -ENOSPC; + + /* + * We have to COW and we have already reserved data space before, + * so now we reserve only metadata. + */ + ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len, len, + false); + if (ret < 0) + goto out; + space_reserved = true; + + em = btrfs_new_extent_direct(BTRFS_I(inode), start, len); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + goto out; + } + *map = em; + len = min(len, em->len - (start - em->start)); + if (len < prev_len) + btrfs_delalloc_release_metadata(BTRFS_I(inode), + prev_len - len, true); + } + + /* + * We have created our ordered extent, so we can now release our reservation + * for an outstanding extent. + */ + btrfs_delalloc_release_extents(BTRFS_I(inode), prev_len); /* * Need to update the i_size under the extent lock so buffered * readers will get the updated i_size when we unlock. */ - if (!dio_data->overwrite && start + len > i_size_read(inode)) + if (start + len > i_size_read(inode)) i_size_write(inode, start + len); - - WARN_ON(dio_data->reserve < len); - dio_data->reserve -= len; - dio_data->unsubmitted_oe_range_end = start + len; - current->journal_info = dio_data; out: + if (ret && space_reserved) { + btrfs_delalloc_release_extents(BTRFS_I(inode), len); + btrfs_delalloc_release_metadata(BTRFS_I(inode), len, true); + } return ret; } -static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) +static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start, + loff_t length, unsigned int flags, struct iomap *iomap, + struct iomap *srcmap) { + struct iomap_iter *iter = container_of(iomap, struct iomap_iter, iomap); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct extent_map *em; struct extent_state *cached_state = NULL; - struct btrfs_dio_data *dio_data = NULL; - u64 start = iblock << inode->i_blkbits; + struct btrfs_dio_data *dio_data = iter->private; u64 lockstart, lockend; - u64 len = bh_result->b_size; + const bool write = !!(flags & IOMAP_WRITE); int ret = 0; + u64 len = length; + const u64 data_alloc_len = length; + bool unlock_extents = false; - if (!create) - len = min_t(u64, len, fs_info->sectorsize); + /* + * We could potentially fault if we have a buffer > PAGE_SIZE, and if + * we're NOWAIT we may submit a bio for a partial range and return + * EIOCBQUEUED, which would result in an errant short read. + * + * The best way to handle this would be to allow for partial completions + * of iocb's, so we could submit the partial bio, return and fault in + * the rest of the pages, and then submit the io for the rest of the + * range. However we don't have that currently, so simply return + * -EAGAIN at this point so that the normal path is used. + */ + if (!write && (flags & IOMAP_NOWAIT) && length > PAGE_SIZE) + return -EAGAIN; + + /* + * Cap the size of reads to that usually seen in buffered I/O as we need + * to allocate a contiguous array for the checksums. + */ + if (!write) + len = min_t(u64, len, fs_info->sectorsize * BTRFS_MAX_BIO_SECTORS); lockstart = start; lockend = start + len - 1; - if (current->journal_info) { - /* - * Need to pull our outstanding extents and set journal_info to NULL so - * that anything that needs to check if there's a transaction doesn't get - * confused. - */ - dio_data = current->journal_info; - current->journal_info = NULL; + /* + * iomap_dio_rw() only does filemap_write_and_wait_range(), which isn't + * enough if we've written compressed pages to this area, so we need to + * flush the dirty pages again to make absolutely sure that any + * outstanding dirty pages are on disk - the first flush only starts + * compression on the data, while keeping the pages locked, so by the + * time the second flush returns we know bios for the compressed pages + * were submitted and finished, and the pages no longer under writeback. + * + * If we have a NOWAIT request and we have any pages in the range that + * are locked, likely due to compression still in progress, we don't want + * to block on page locks. We also don't want to block on pages marked as + * dirty or under writeback (same as for the non-compression case). + * iomap_dio_rw() did the same check, but after that and before we got + * here, mmap'ed writes may have happened or buffered reads started + * (readpage() and readahead(), which lock pages), as we haven't locked + * the file range yet. + */ + if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, + &BTRFS_I(inode)->runtime_flags)) { + if (flags & IOMAP_NOWAIT) { + if (filemap_range_needs_writeback(inode->i_mapping, + lockstart, lockend)) + return -EAGAIN; + } else { + ret = filemap_fdatawrite_range(inode->i_mapping, start, + start + length - 1); + if (ret) + return ret; + } + } + + memset(dio_data, 0, sizeof(*dio_data)); + + /* + * We always try to allocate data space and must do it before locking + * the file range, to avoid deadlocks with concurrent writes to the same + * range if the range has several extents and the writes don't expand the + * current i_size (the inode lock is taken in shared mode). If we fail to + * allocate data space here we continue and later, after locking the + * file range, we fail with ENOSPC only if we figure out we can not do a + * NOCOW write. + */ + if (write && !(flags & IOMAP_NOWAIT)) { + ret = btrfs_check_data_free_space(BTRFS_I(inode), + &dio_data->data_reserved, + start, data_alloc_len, false); + if (!ret) + dio_data->data_space_reserved = true; + else if (ret && !(BTRFS_I(inode)->flags & + (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC))) + goto err; } /* * If this errors out it's because we couldn't invalidate pagecache for - * this range and we need to fallback to buffered. + * this range and we need to fallback to buffered IO, or we are doing a + * NOWAIT read/write and we need to block. */ - if (lock_extent_direct(inode, lockstart, lockend, &cached_state, - create)) { - ret = -ENOTBLK; + ret = lock_extent_direct(inode, lockstart, lockend, &cached_state, flags); + if (ret < 0) goto err; - } em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len); if (IS_ERR(em)) { @@ -7262,1128 +7643,737 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags) || em->block_start == EXTENT_MAP_INLINE) { free_extent_map(em); - ret = -ENOTBLK; + /* + * If we are in a NOWAIT context, return -EAGAIN in order to + * fallback to buffered IO. This is not only because we can + * block with buffered IO (no support for NOWAIT semantics at + * the moment) but also to avoid returning short reads to user + * space - this happens if we were able to read some data from + * previous non-compressed extents and then when we fallback to + * buffered IO, at btrfs_file_read_iter() by calling + * filemap_read(), we fail to fault in pages for the read buffer, + * in which case filemap_read() returns a short read (the number + * of bytes previously read is > 0, so it does not return -EFAULT). + */ + ret = (flags & IOMAP_NOWAIT) ? -EAGAIN : -ENOTBLK; goto unlock_err; } - if (create) { - ret = btrfs_get_blocks_direct_write(&em, bh_result, inode, - dio_data, start, len); + len = min(len, em->len - (start - em->start)); + + /* + * If we have a NOWAIT request and the range contains multiple extents + * (or a mix of extents and holes), then we return -EAGAIN to make the + * caller fallback to a context where it can do a blocking (without + * NOWAIT) request. This way we avoid doing partial IO and returning + * success to the caller, which is not optimal for writes and for reads + * it can result in unexpected behaviour for an application. + * + * When doing a read, because we use IOMAP_DIO_PARTIAL when calling + * iomap_dio_rw(), we can end up returning less data then what the caller + * asked for, resulting in an unexpected, and incorrect, short read. + * That is, the caller asked to read N bytes and we return less than that, + * which is wrong unless we are crossing EOF. This happens if we get a + * page fault error when trying to fault in pages for the buffer that is + * associated to the struct iov_iter passed to iomap_dio_rw(), and we + * have previously submitted bios for other extents in the range, in + * which case iomap_dio_rw() may return us EIOCBQUEUED if not all of + * those bios have completed by the time we get the page fault error, + * which we return back to our caller - we should only return EIOCBQUEUED + * after we have submitted bios for all the extents in the range. + */ + if ((flags & IOMAP_NOWAIT) && len < length) { + free_extent_map(em); + ret = -EAGAIN; + goto unlock_err; + } + + if (write) { + ret = btrfs_get_blocks_direct_write(&em, inode, dio_data, + start, len, flags); if (ret < 0) goto unlock_err; + unlock_extents = true; + /* Recalc len in case the new em is smaller than requested */ + len = min(len, em->len - (start - em->start)); + if (dio_data->data_space_reserved) { + u64 release_offset; + u64 release_len = 0; + + if (dio_data->nocow_done) { + release_offset = start; + release_len = data_alloc_len; + } else if (len < data_alloc_len) { + release_offset = start + len; + release_len = data_alloc_len - len; + } - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, - lockend, &cached_state); - } else { - ret = btrfs_get_blocks_direct_read(em, bh_result, inode, - start, len); - /* Can be negative only if we read from a hole */ - if (ret < 0) { - ret = 0; - free_extent_map(em); - goto unlock_err; + if (release_len > 0) + btrfs_free_reserved_data_space(BTRFS_I(inode), + dio_data->data_reserved, + release_offset, + release_len); } + } else { /* * We need to unlock only the end area that we aren't using. * The rest is going to be unlocked by the endio routine. */ - lockstart = start + bh_result->b_size; - if (lockstart < lockend) { - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - lockstart, lockend, &cached_state); - } else { - free_extent_state(cached_state); - } + lockstart = start + len; + if (lockstart < lockend) + unlock_extents = true; } + if (unlock_extents) + unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + &cached_state); + else + free_extent_state(cached_state); + + /* + * Translate extent map information to iomap. + * We trim the extents (and move the addr) even though iomap code does + * that, since we have locked only the parts we are performing I/O in. + */ + if ((em->block_start == EXTENT_MAP_HOLE) || + (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) && !write)) { + iomap->addr = IOMAP_NULL_ADDR; + iomap->type = IOMAP_HOLE; + } else { + iomap->addr = em->block_start + (start - em->start); + iomap->type = IOMAP_MAPPED; + } + iomap->offset = start; + iomap->bdev = fs_info->fs_devices->latest_dev->bdev; + iomap->length = len; + + if (write && btrfs_use_zone_append(BTRFS_I(inode), em->block_start)) + iomap->flags |= IOMAP_F_ZONE_APPEND; + free_extent_map(em); return 0; unlock_err: - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, - &cached_state); + unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + &cached_state); err: - if (dio_data) - current->journal_info = dio_data; - return ret; -} - -static inline blk_status_t submit_dio_repair_bio(struct inode *inode, - struct bio *bio, - int mirror_num) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - blk_status_t ret; - - BUG_ON(bio_op(bio) == REQ_OP_WRITE); - - ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DIO_REPAIR); - if (ret) - return ret; - - ret = btrfs_map_bio(fs_info, bio, mirror_num); + if (dio_data->data_space_reserved) { + btrfs_free_reserved_data_space(BTRFS_I(inode), + dio_data->data_reserved, + start, data_alloc_len); + extent_changeset_free(dio_data->data_reserved); + } return ret; } -static int btrfs_check_dio_repairable(struct inode *inode, - struct bio *failed_bio, - struct io_failure_record *failrec, - int failed_mirror) +static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length, + ssize_t written, unsigned int flags, struct iomap *iomap) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int num_copies; - - num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len); - if (num_copies == 1) { - /* - * we only have a single copy of the data, so don't bother with - * all the retry and error correction code that follows. no - * matter what the error is, it is very likely to persist. - */ - btrfs_debug(fs_info, - "Check DIO Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d", - num_copies, failrec->this_mirror, failed_mirror); - return 0; - } - - failrec->failed_mirror = failed_mirror; - failrec->this_mirror++; - if (failrec->this_mirror == failed_mirror) - failrec->this_mirror++; + struct iomap_iter *iter = container_of(iomap, struct iomap_iter, iomap); + struct btrfs_dio_data *dio_data = iter->private; + size_t submitted = dio_data->submitted; + const bool write = !!(flags & IOMAP_WRITE); + int ret = 0; - if (failrec->this_mirror > num_copies) { - btrfs_debug(fs_info, - "Check DIO Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d", - num_copies, failrec->this_mirror, failed_mirror); + if (!write && (iomap->type == IOMAP_HOLE)) { + /* If reading from a hole, unlock and return */ + unlock_extent(&BTRFS_I(inode)->io_tree, pos, pos + length - 1, + NULL); return 0; } - return 1; -} - -static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio, - struct page *page, unsigned int pgoff, - u64 start, u64 end, int failed_mirror, - bio_end_io_t *repair_endio, void *repair_arg) -{ - struct io_failure_record *failrec; - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree; - struct bio *bio; - int isector; - unsigned int read_mode = 0; - int segs; - int ret; - blk_status_t status; - struct bio_vec bvec; - - BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE); - - ret = btrfs_get_io_failure_record(inode, start, end, &failrec); - if (ret) - return errno_to_blk_status(ret); - - ret = btrfs_check_dio_repairable(inode, failed_bio, failrec, - failed_mirror); - if (!ret) { - free_io_failure(failure_tree, io_tree, failrec); - return BLK_STS_IOERR; - } - - segs = bio_segments(failed_bio); - bio_get_first_bvec(failed_bio, &bvec); - if (segs > 1 || - (bvec.bv_len > btrfs_inode_sectorsize(inode))) - read_mode |= REQ_FAILFAST_DEV; - - isector = start - btrfs_io_bio(failed_bio)->logical; - isector >>= inode->i_sb->s_blocksize_bits; - bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page, - pgoff, isector, repair_endio, repair_arg); - bio->bi_opf = REQ_OP_READ | read_mode; - - btrfs_debug(BTRFS_I(inode)->root->fs_info, - "repair DIO read error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d", - read_mode, failrec->this_mirror, failrec->in_validation); - - status = submit_dio_repair_bio(inode, bio, failrec->this_mirror); - if (status) { - free_io_failure(failure_tree, io_tree, failrec); - bio_put(bio); + if (submitted < length) { + pos += submitted; + length -= submitted; + if (write) + btrfs_mark_ordered_io_finished(BTRFS_I(inode), NULL, + pos, length, false); + else + unlock_extent(&BTRFS_I(inode)->io_tree, pos, + pos + length - 1, NULL); + ret = -ENOTBLK; } - return status; -} - -struct btrfs_retry_complete { - struct completion done; - struct inode *inode; - u64 start; - int uptodate; -}; - -static void btrfs_retry_endio_nocsum(struct bio *bio) -{ - struct btrfs_retry_complete *done = bio->bi_private; - struct inode *inode = done->inode; - struct bio_vec *bvec; - struct extent_io_tree *io_tree, *failure_tree; - struct bvec_iter_all iter_all; - - if (bio->bi_status) - goto end; - - ASSERT(bio->bi_vcnt == 1); - io_tree = &BTRFS_I(inode)->io_tree; - failure_tree = &BTRFS_I(inode)->io_failure_tree; - ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(inode)); - - done->uptodate = 1; - ASSERT(!bio_flagged(bio, BIO_CLONED)); - bio_for_each_segment_all(bvec, bio, iter_all) - clean_io_failure(BTRFS_I(inode)->root->fs_info, failure_tree, - io_tree, done->start, bvec->bv_page, - btrfs_ino(BTRFS_I(inode)), 0); -end: - complete(&done->done); - bio_put(bio); + if (write) + extent_changeset_free(dio_data->data_reserved); + return ret; } -static blk_status_t __btrfs_correct_data_nocsum(struct inode *inode, - struct btrfs_io_bio *io_bio) +static void btrfs_dio_private_put(struct btrfs_dio_private *dip) { - struct btrfs_fs_info *fs_info; - struct bio_vec bvec; - struct bvec_iter iter; - struct btrfs_retry_complete done; - u64 start; - unsigned int pgoff; - u32 sectorsize; - int nr_sectors; - blk_status_t ret; - blk_status_t err = BLK_STS_OK; - - fs_info = BTRFS_I(inode)->root->fs_info; - sectorsize = fs_info->sectorsize; - - start = io_bio->logical; - done.inode = inode; - io_bio->bio.bi_iter = io_bio->iter; - - bio_for_each_segment(bvec, &io_bio->bio, iter) { - nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len); - pgoff = bvec.bv_offset; - -next_block_or_try_again: - done.uptodate = 0; - done.start = start; - init_completion(&done.done); - - ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page, - pgoff, start, start + sectorsize - 1, - io_bio->mirror_num, - btrfs_retry_endio_nocsum, &done); - if (ret) { - err = ret; - goto next; - } - - wait_for_completion_io(&done.done); - - if (!done.uptodate) { - /* We might have another mirror, so try again */ - goto next_block_or_try_again; - } - -next: - start += sectorsize; + /* + * This implies a barrier so that stores to dio_bio->bi_status before + * this and loads of dio_bio->bi_status after this are fully ordered. + */ + if (!refcount_dec_and_test(&dip->refs)) + return; - nr_sectors--; - if (nr_sectors) { - pgoff += sectorsize; - ASSERT(pgoff < PAGE_SIZE); - goto next_block_or_try_again; - } + if (btrfs_op(&dip->bio) == BTRFS_MAP_WRITE) { + btrfs_mark_ordered_io_finished(BTRFS_I(dip->inode), NULL, + dip->file_offset, dip->bytes, + !dip->bio.bi_status); + } else { + unlock_extent(&BTRFS_I(dip->inode)->io_tree, + dip->file_offset, + dip->file_offset + dip->bytes - 1, NULL); } - return err; + kfree(dip->csums); + bio_endio(&dip->bio); } -static void btrfs_retry_endio(struct bio *bio) +static void submit_dio_repair_bio(struct inode *inode, struct bio *bio, + int mirror_num, + enum btrfs_compression_type compress_type) { - struct btrfs_retry_complete *done = bio->bi_private; - struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); - struct extent_io_tree *io_tree, *failure_tree; - struct inode *inode = done->inode; - struct bio_vec *bvec; - int uptodate; - int ret; - int i = 0; - struct bvec_iter_all iter_all; - - if (bio->bi_status) - goto end; - - uptodate = 1; - - ASSERT(bio->bi_vcnt == 1); - ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(done->inode)); - - io_tree = &BTRFS_I(inode)->io_tree; - failure_tree = &BTRFS_I(inode)->io_failure_tree; + struct btrfs_dio_private *dip = btrfs_bio(bio)->private; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - ASSERT(!bio_flagged(bio, BIO_CLONED)); - bio_for_each_segment_all(bvec, bio, iter_all) { - ret = __readpage_endio_check(inode, io_bio, i, bvec->bv_page, - bvec->bv_offset, done->start, - bvec->bv_len); - if (!ret) - clean_io_failure(BTRFS_I(inode)->root->fs_info, - failure_tree, io_tree, done->start, - bvec->bv_page, - btrfs_ino(BTRFS_I(inode)), - bvec->bv_offset); - else - uptodate = 0; - i++; - } + BUG_ON(bio_op(bio) == REQ_OP_WRITE); - done->uptodate = uptodate; -end: - complete(&done->done); - bio_put(bio); + refcount_inc(&dip->refs); + btrfs_submit_bio(fs_info, bio, mirror_num); } -static blk_status_t __btrfs_subio_endio_read(struct inode *inode, - struct btrfs_io_bio *io_bio, blk_status_t err) +static blk_status_t btrfs_check_read_dio_bio(struct btrfs_dio_private *dip, + struct btrfs_bio *bbio, + const bool uptodate) { - struct btrfs_fs_info *fs_info; - struct bio_vec bvec; + struct inode *inode = dip->inode; + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM); + blk_status_t err = BLK_STS_OK; struct bvec_iter iter; - struct btrfs_retry_complete done; - u64 start; - u64 offset = 0; - u32 sectorsize; - int nr_sectors; - unsigned int pgoff; - int csum_pos; - bool uptodate = (err == 0); - int ret; - blk_status_t status; - - fs_info = BTRFS_I(inode)->root->fs_info; - sectorsize = fs_info->sectorsize; - - err = BLK_STS_OK; - start = io_bio->logical; - done.inode = inode; - io_bio->bio.bi_iter = io_bio->iter; + struct bio_vec bv; + u32 offset; - bio_for_each_segment(bvec, &io_bio->bio, iter) { - nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len); + btrfs_bio_for_each_sector(fs_info, bv, bbio, iter, offset) { + u64 start = bbio->file_offset + offset; - pgoff = bvec.bv_offset; -next_block: - if (uptodate) { - csum_pos = BTRFS_BYTES_TO_BLKS(fs_info, offset); - ret = __readpage_endio_check(inode, io_bio, csum_pos, - bvec.bv_page, pgoff, start, sectorsize); - if (likely(!ret)) - goto next; - } -try_again: - done.uptodate = 0; - done.start = start; - init_completion(&done.done); - - status = dio_read_error(inode, &io_bio->bio, bvec.bv_page, - pgoff, start, start + sectorsize - 1, - io_bio->mirror_num, btrfs_retry_endio, - &done); - if (status) { - err = status; - goto next; - } - - wait_for_completion_io(&done.done); - - if (!done.uptodate) { - /* We might have another mirror, so try again */ - goto try_again; - } -next: - offset += sectorsize; - start += sectorsize; - - ASSERT(nr_sectors); + if (uptodate && + (!csum || !btrfs_check_data_csum(inode, bbio, offset, bv.bv_page, + bv.bv_offset))) { + btrfs_clean_io_failure(BTRFS_I(inode), start, + bv.bv_page, bv.bv_offset); + } else { + int ret; - nr_sectors--; - if (nr_sectors) { - pgoff += sectorsize; - ASSERT(pgoff < PAGE_SIZE); - goto next_block; + ret = btrfs_repair_one_sector(inode, bbio, offset, + bv.bv_page, bv.bv_offset, + submit_dio_repair_bio); + if (ret) + err = errno_to_blk_status(ret); } } return err; } -static blk_status_t btrfs_subio_endio_read(struct inode *inode, - struct btrfs_io_bio *io_bio, blk_status_t err) -{ - bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - - if (skip_csum) { - if (unlikely(err)) - return __btrfs_correct_data_nocsum(inode, io_bio); - else - return BLK_STS_OK; - } else { - return __btrfs_subio_endio_read(inode, io_bio, err); - } -} - -static void btrfs_endio_direct_read(struct bio *bio) -{ - struct btrfs_dio_private *dip = bio->bi_private; - struct inode *inode = dip->inode; - struct bio *dio_bio; - struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); - blk_status_t err = bio->bi_status; - - if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED) - err = btrfs_subio_endio_read(inode, io_bio, err); - - unlock_extent(&BTRFS_I(inode)->io_tree, dip->logical_offset, - dip->logical_offset + dip->bytes - 1); - dio_bio = dip->dio_bio; - - kfree(dip); - - dio_bio->bi_status = err; - dio_end_io(dio_bio); - btrfs_io_bio_free_csum(io_bio); - bio_put(bio); -} - -static void __endio_write_update_ordered(struct inode *inode, - const u64 offset, const u64 bytes, - const bool uptodate) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_ordered_extent *ordered = NULL; - struct btrfs_workqueue *wq; - u64 ordered_offset = offset; - u64 ordered_bytes = bytes; - u64 last_offset; - - if (btrfs_is_free_space_inode(BTRFS_I(inode))) - wq = fs_info->endio_freespace_worker; - else - wq = fs_info->endio_write_workers; - - while (ordered_offset < offset + bytes) { - last_offset = ordered_offset; - if (btrfs_dec_test_first_ordered_pending(inode, &ordered, - &ordered_offset, - ordered_bytes, - uptodate)) { - btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, - NULL); - btrfs_queue_work(wq, &ordered->work); - } - /* - * If btrfs_dec_test_ordered_pending does not find any ordered - * extent in the range, we can exit. - */ - if (ordered_offset == last_offset) - return; - /* - * Our bio might span multiple ordered extents. In this case - * we keep going until we have accounted the whole dio. - */ - if (ordered_offset < offset + bytes) { - ordered_bytes = offset + bytes - ordered_offset; - ordered = NULL; - } - } -} - -static void btrfs_endio_direct_write(struct bio *bio) +static blk_status_t btrfs_submit_bio_start_direct_io(struct inode *inode, + struct bio *bio, + u64 dio_file_offset) { - struct btrfs_dio_private *dip = bio->bi_private; - struct bio *dio_bio = dip->dio_bio; - - __endio_write_update_ordered(dip->inode, dip->logical_offset, - dip->bytes, !bio->bi_status); - - kfree(dip); - - dio_bio->bi_status = bio->bi_status; - dio_end_io(dio_bio); - bio_put(bio); -} - -static blk_status_t btrfs_submit_bio_start_direct_io(void *private_data, - struct bio *bio, u64 offset) -{ - struct inode *inode = private_data; - blk_status_t ret; - ret = btrfs_csum_one_bio(inode, bio, offset, 1); - BUG_ON(ret); /* -ENOMEM */ - return 0; + return btrfs_csum_one_bio(BTRFS_I(inode), bio, dio_file_offset, false); } -static void btrfs_end_dio_bio(struct bio *bio) +static void btrfs_end_dio_bio(struct btrfs_bio *bbio) { - struct btrfs_dio_private *dip = bio->bi_private; + struct btrfs_dio_private *dip = bbio->private; + struct bio *bio = &bbio->bio; blk_status_t err = bio->bi_status; if (err) btrfs_warn(BTRFS_I(dip->inode)->root->fs_info, "direct IO failed ino %llu rw %d,%u sector %#Lx len %u err no %d", btrfs_ino(BTRFS_I(dip->inode)), bio_op(bio), - bio->bi_opf, - (unsigned long long)bio->bi_iter.bi_sector, + bio->bi_opf, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, err); - if (dip->subio_endio) - err = dip->subio_endio(dip->inode, btrfs_io_bio(bio), err); + if (bio_op(bio) == REQ_OP_READ) + err = btrfs_check_read_dio_bio(dip, bbio, !err); - if (err) { - /* - * We want to perceive the errors flag being set before - * decrementing the reference count. We don't need a barrier - * since atomic operations with a return value are fully - * ordered as per atomic_t.txt - */ - dip->errors = 1; - } + if (err) + dip->bio.bi_status = err; - /* if there are more bios still pending for this dio, just exit */ - if (!atomic_dec_and_test(&dip->pending_bios)) - goto out; + btrfs_record_physical_zoned(dip->inode, bbio->file_offset, bio); - if (dip->errors) { - bio_io_error(dip->orig_bio); - } else { - dip->dio_bio->bi_status = BLK_STS_OK; - bio_endio(dip->orig_bio); - } -out: bio_put(bio); + btrfs_dio_private_put(dip); } -static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode, - struct btrfs_dio_private *dip, - struct bio *bio, - u64 file_offset) -{ - struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); - struct btrfs_io_bio *orig_io_bio = btrfs_io_bio(dip->orig_bio); - u16 csum_size; - blk_status_t ret; - - /* - * We load all the csum data we need when we submit - * the first bio to reduce the csum tree search and - * contention. - */ - if (dip->logical_offset == file_offset) { - ret = btrfs_lookup_bio_sums(inode, dip->orig_bio, file_offset, - NULL); - if (ret) - return ret; - } - - if (bio == dip->orig_bio) - return 0; - - file_offset -= dip->logical_offset; - file_offset >>= inode->i_sb->s_blocksize_bits; - csum_size = btrfs_super_csum_size(btrfs_sb(inode->i_sb)->super_copy); - io_bio->csum = orig_io_bio->csum + csum_size * file_offset; - - return 0; -} - -static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio, - struct inode *inode, u64 file_offset, int async_submit) +static void btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, + u64 file_offset, 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_dio_private *dip = btrfs_bio(bio)->private; blk_status_t ret; - /* Check btrfs_submit_bio_hook() for rules about async submit. */ - if (async_submit) - async_submit = !atomic_read(&BTRFS_I(inode)->sync_writers); - - if (!write) { - ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA); - if (ret) - goto err; - } + /* Save the original iter for read repair */ + if (btrfs_op(bio) == BTRFS_MAP_READ) + btrfs_bio(bio)->iter = bio->bi_iter; if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) goto map; - if (write && async_submit) { - ret = btrfs_wq_submit_bio(fs_info, bio, 0, 0, - file_offset, inode, - btrfs_submit_bio_start_direct_io); - goto err; - } else if (write) { + if (btrfs_op(bio) == BTRFS_MAP_WRITE) { + /* Check btrfs_submit_data_write_bio() for async submit rules */ + if (async_submit && !atomic_read(&BTRFS_I(inode)->sync_writers) && + btrfs_wq_submit_bio(inode, bio, 0, file_offset, + btrfs_submit_bio_start_direct_io)) + return; + /* * If we aren't doing async submit, calculate the csum of the * bio now. */ - ret = btrfs_csum_one_bio(inode, bio, file_offset, 1); - if (ret) - goto err; + ret = btrfs_csum_one_bio(BTRFS_I(inode), bio, file_offset, false); + if (ret) { + btrfs_bio_end_io(btrfs_bio(bio), ret); + return; + } } else { - ret = btrfs_lookup_and_bind_dio_csum(inode, dip, bio, - file_offset); - if (ret) - goto err; + btrfs_bio(bio)->csum = btrfs_csum_ptr(fs_info, dip->csums, + file_offset - dip->file_offset); } map: - ret = btrfs_map_bio(fs_info, bio, 0); -err: - return ret; + btrfs_submit_bio(fs_info, bio, 0); } -static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip) +static void btrfs_submit_direct(const struct iomap_iter *iter, + struct bio *dio_bio, loff_t file_offset) { - struct inode *inode = dip->inode; + struct btrfs_dio_private *dip = + container_of(dio_bio, struct btrfs_dio_private, bio); + struct inode *inode = iter->inode; + const bool write = (btrfs_op(dio_bio) == BTRFS_MAP_WRITE); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + const bool raid56 = (btrfs_data_alloc_profile(fs_info) & + BTRFS_BLOCK_GROUP_RAID56_MASK); struct bio *bio; - struct bio *orig_bio = dip->orig_bio; - u64 start_sector = orig_bio->bi_iter.bi_sector; - u64 file_offset = dip->logical_offset; + u64 start_sector; int async_submit = 0; u64 submit_len; - int clone_offset = 0; - int clone_len; + u64 clone_offset = 0; + u64 clone_len; + u64 logical; int ret; blk_status_t status; struct btrfs_io_geometry geom; + struct btrfs_dio_data *dio_data = iter->private; + struct extent_map *em = NULL; - submit_len = orig_bio->bi_iter.bi_size; - ret = btrfs_get_io_geometry(fs_info, btrfs_op(orig_bio), - start_sector << 9, submit_len, &geom); - if (ret) - return -EIO; + dip->inode = inode; + dip->file_offset = file_offset; + dip->bytes = dio_bio->bi_iter.bi_size; + refcount_set(&dip->refs, 1); + dip->csums = NULL; - if (geom.len >= submit_len) { - bio = orig_bio; - dip->flags |= BTRFS_DIO_ORIG_BIO_SUBMITTED; - goto submit; + if (!write && !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) { + unsigned int nr_sectors = + (dio_bio->bi_iter.bi_size >> fs_info->sectorsize_bits); + + /* + * Load the csums up front to reduce csum tree searches and + * contention when submitting bios. + */ + status = BLK_STS_RESOURCE; + dip->csums = kcalloc(nr_sectors, fs_info->csum_size, GFP_NOFS); + if (!dip->csums) + goto out_err; + + status = btrfs_lookup_bio_sums(inode, dio_bio, dip->csums); + if (status != BLK_STS_OK) + goto out_err; } - /* async crcs make it difficult to collect full stripe writes. */ - if (btrfs_data_alloc_profile(fs_info) & BTRFS_BLOCK_GROUP_RAID56_MASK) - async_submit = 0; - else - async_submit = 1; + start_sector = dio_bio->bi_iter.bi_sector; + submit_len = dio_bio->bi_iter.bi_size; - /* bio split */ - ASSERT(geom.len <= INT_MAX); - atomic_inc(&dip->pending_bios); do { - clone_len = min_t(int, submit_len, geom.len); + logical = start_sector << 9; + em = btrfs_get_chunk_map(fs_info, logical, submit_len); + if (IS_ERR(em)) { + status = errno_to_blk_status(PTR_ERR(em)); + em = NULL; + goto out_err_em; + } + ret = btrfs_get_io_geometry(fs_info, em, btrfs_op(dio_bio), + logical, &geom); + if (ret) { + status = errno_to_blk_status(ret); + goto out_err_em; + } + + clone_len = min(submit_len, geom.len); + ASSERT(clone_len <= UINT_MAX); /* * This will never fail as it's passing GPF_NOFS and * the allocation is backed by btrfs_bioset. */ - bio = btrfs_bio_clone_partial(orig_bio, clone_offset, - clone_len); - bio->bi_private = dip; - bio->bi_end_io = btrfs_end_dio_bio; - btrfs_io_bio(bio)->logical = file_offset; + bio = btrfs_bio_clone_partial(dio_bio, clone_offset, clone_len, + btrfs_end_dio_bio, dip); + btrfs_bio(bio)->file_offset = file_offset; + + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + status = extract_ordered_extent(BTRFS_I(inode), bio, + file_offset); + if (status) { + bio_put(bio); + goto out_err; + } + } ASSERT(submit_len >= clone_len); submit_len -= clone_len; - if (submit_len == 0) - break; /* * Increase the count before we submit the bio so we know * the end IO handler won't happen before we increase the * count. Otherwise, the dip might get freed before we're * done setting it up. + * + * We transfer the initial reference to the last bio, so we + * don't need to increment the reference count for the last one. */ - atomic_inc(&dip->pending_bios); - - status = btrfs_submit_dio_bio(bio, inode, file_offset, - async_submit); - if (status) { - bio_put(bio); - atomic_dec(&dip->pending_bios); - goto out_err; + if (submit_len > 0) { + refcount_inc(&dip->refs); + /* + * If we are submitting more than one bio, submit them + * all asynchronously. The exception is RAID 5 or 6, as + * asynchronous checksums make it difficult to collect + * full stripe writes. + */ + if (!raid56) + async_submit = 1; } + btrfs_submit_dio_bio(bio, inode, file_offset, async_submit); + + dio_data->submitted += clone_len; clone_offset += clone_len; start_sector += clone_len >> 9; file_offset += clone_len; - ret = btrfs_get_io_geometry(fs_info, btrfs_op(orig_bio), - start_sector << 9, submit_len, &geom); - if (ret) - goto out_err; + free_extent_map(em); } while (submit_len > 0); + return; -submit: - status = btrfs_submit_dio_bio(bio, inode, file_offset, async_submit); - if (!status) - return 0; - - bio_put(bio); +out_err_em: + free_extent_map(em); out_err: - dip->errors = 1; - /* - * Before atomic variable goto zero, we must make sure dip->errors is - * perceived to be set. This ordering is ensured by the fact that an - * atomic operations with a return value are fully ordered as per - * atomic_t.txt - */ - if (atomic_dec_and_test(&dip->pending_bios)) - bio_io_error(dip->orig_bio); - - /* bio_end_io() will handle error, so we needn't return it */ - return 0; + dio_bio->bi_status = status; + btrfs_dio_private_put(dip); } -static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode, - loff_t file_offset) -{ - struct btrfs_dio_private *dip = NULL; - struct bio *bio = NULL; - struct btrfs_io_bio *io_bio; - bool write = (bio_op(dio_bio) == REQ_OP_WRITE); - int ret = 0; - - bio = btrfs_bio_clone(dio_bio); - - dip = kzalloc(sizeof(*dip), GFP_NOFS); - if (!dip) { - ret = -ENOMEM; - goto free_ordered; - } - - dip->private = dio_bio->bi_private; - dip->inode = inode; - dip->logical_offset = file_offset; - dip->bytes = dio_bio->bi_iter.bi_size; - dip->disk_bytenr = (u64)dio_bio->bi_iter.bi_sector << 9; - bio->bi_private = dip; - dip->orig_bio = bio; - dip->dio_bio = dio_bio; - atomic_set(&dip->pending_bios, 0); - io_bio = btrfs_io_bio(bio); - io_bio->logical = file_offset; - - if (write) { - bio->bi_end_io = btrfs_endio_direct_write; - } else { - bio->bi_end_io = btrfs_endio_direct_read; - dip->subio_endio = btrfs_subio_endio_read; - } - - /* - * Reset the range for unsubmitted ordered extents (to a 0 length range) - * even if we fail to submit a bio, because in such case we do the - * corresponding error handling below and it must not be done a second - * time by btrfs_direct_IO(). - */ - if (write) { - struct btrfs_dio_data *dio_data = current->journal_info; - - dio_data->unsubmitted_oe_range_end = dip->logical_offset + - dip->bytes; - dio_data->unsubmitted_oe_range_start = - dio_data->unsubmitted_oe_range_end; - } - - ret = btrfs_submit_direct_hook(dip); - if (!ret) - return; +static const struct iomap_ops btrfs_dio_iomap_ops = { + .iomap_begin = btrfs_dio_iomap_begin, + .iomap_end = btrfs_dio_iomap_end, +}; - btrfs_io_bio_free_csum(io_bio); +static const struct iomap_dio_ops btrfs_dio_ops = { + .submit_io = btrfs_submit_direct, + .bio_set = &btrfs_dio_bioset, +}; -free_ordered: - /* - * If we arrived here it means either we failed to submit the dip - * or we either failed to clone the dio_bio or failed to allocate the - * dip. If we cloned the dio_bio and allocated the dip, we can just - * call bio_endio against our io_bio so that we get proper resource - * cleanup if we fail to submit the dip, otherwise, we must do the - * same as btrfs_endio_direct_[write|read] because we can't call these - * callbacks - they require an allocated dip and a clone of dio_bio. - */ - if (bio && dip) { - bio_io_error(bio); - /* - * The end io callbacks free our dip, do the final put on bio - * and all the cleanup and final put for dio_bio (through - * dio_end_io()). - */ - dip = NULL; - bio = NULL; - } else { - if (write) - __endio_write_update_ordered(inode, - file_offset, - dio_bio->bi_iter.bi_size, - false); - else - unlock_extent(&BTRFS_I(inode)->io_tree, file_offset, - file_offset + dio_bio->bi_iter.bi_size - 1); +ssize_t btrfs_dio_read(struct kiocb *iocb, struct iov_iter *iter, size_t done_before) +{ + struct btrfs_dio_data data; - dio_bio->bi_status = BLK_STS_IOERR; - /* - * Releases and cleans up our dio_bio, no need to bio_put() - * nor bio_endio()/bio_io_error() against dio_bio. - */ - dio_end_io(dio_bio); - } - if (bio) - bio_put(bio); - kfree(dip); + return iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops, &btrfs_dio_ops, + IOMAP_DIO_PARTIAL, &data, done_before); } -static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info, - const struct iov_iter *iter, loff_t offset) +struct iomap_dio *btrfs_dio_write(struct kiocb *iocb, struct iov_iter *iter, + size_t done_before) { - int seg; - int i; - unsigned int blocksize_mask = fs_info->sectorsize - 1; - ssize_t retval = -EINVAL; + struct btrfs_dio_data data; - if (offset & blocksize_mask) - goto out; - - if (iov_iter_alignment(iter) & blocksize_mask) - goto out; - - /* If this is a write we don't need to check anymore */ - if (iov_iter_rw(iter) != READ || !iter_is_iovec(iter)) - return 0; - /* - * Check to make sure we don't have duplicate iov_base's in this - * iovec, if so return EINVAL, otherwise we'll get csum errors - * when reading back. - */ - for (seg = 0; seg < iter->nr_segs; seg++) { - for (i = seg + 1; i < iter->nr_segs; i++) { - if (iter->iov[seg].iov_base == iter->iov[i].iov_base) - goto out; - } - } - retval = 0; -out: - return retval; + return __iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops, &btrfs_dio_ops, + IOMAP_DIO_PARTIAL, &data, done_before); } -static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) +static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) { - struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_dio_data dio_data = { 0 }; - struct extent_changeset *data_reserved = NULL; - loff_t offset = iocb->ki_pos; - size_t count = 0; - int flags = 0; - bool wakeup = true; - bool relock = false; - ssize_t ret; - - if (check_direct_IO(fs_info, iter, offset)) - return 0; + int ret; - inode_dio_begin(inode); + ret = fiemap_prep(inode, fieinfo, start, &len, 0); + if (ret) + return ret; /* - * The generic stuff only does filemap_write_and_wait_range, which - * isn't enough if we've written compressed pages to this area, so - * we need to flush the dirty pages again to make absolutely sure - * that any outstanding dirty pages are on disk. + * fiemap_prep() called filemap_write_and_wait() for the whole possible + * file range (0 to LLONG_MAX), but that is not enough if we have + * compression enabled. The first filemap_fdatawrite_range() only kicks + * in the compression of data (in an async thread) and will return + * before the compression is done and writeback is started. A second + * filemap_fdatawrite_range() is needed to wait for the compression to + * complete and writeback to start. We also need to wait for ordered + * extents to complete, because our fiemap implementation uses mainly + * file extent items to list the extents, searching for extent maps + * only for file ranges with holes or prealloc extents to figure out + * if we have delalloc in those ranges. */ - count = iov_iter_count(iter); - if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, - &BTRFS_I(inode)->runtime_flags)) - filemap_fdatawrite_range(inode->i_mapping, offset, - offset + count - 1); - - if (iov_iter_rw(iter) == WRITE) { - /* - * If the write DIO is beyond the EOF, we need update - * the isize, but it is protected by i_mutex. So we can - * not unlock the i_mutex at this case. - */ - if (offset + count <= inode->i_size) { - dio_data.overwrite = 1; - inode_unlock(inode); - relock = true; - } else if (iocb->ki_flags & IOCB_NOWAIT) { - ret = -EAGAIN; - goto out; - } - ret = btrfs_delalloc_reserve_space(inode, &data_reserved, - offset, count); + if (fieinfo->fi_flags & FIEMAP_FLAG_SYNC) { + ret = btrfs_wait_ordered_range(inode, 0, LLONG_MAX); if (ret) - goto out; - - /* - * 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, - 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); - flags = DIO_LOCKING | DIO_SKIP_HOLES; - wakeup = false; - } - - ret = __blockdev_direct_IO(iocb, inode, - 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) - btrfs_delalloc_release_space(inode, data_reserved, - offset, dio_data.reserve, true); - /* - * On error we might have left some ordered extents - * without submitting corresponding bios for them, so - * cleanup them up to avoid other tasks getting them - * and waiting for them to complete forever. - */ - if (dio_data.unsubmitted_oe_range_start < - dio_data.unsubmitted_oe_range_end) - __endio_write_update_ordered(inode, - dio_data.unsubmitted_oe_range_start, - dio_data.unsubmitted_oe_range_end - - dio_data.unsubmitted_oe_range_start, - false); - } else if (ret >= 0 && (size_t)ret < count) - btrfs_delalloc_release_space(inode, data_reserved, - offset, count - (size_t)ret, true); - btrfs_delalloc_release_extents(BTRFS_I(inode), count); + return ret; } -out: - if (wakeup) - inode_dio_end(inode); - if (relock) - inode_lock(inode); - extent_changeset_free(data_reserved); - return ret; + return extent_fiemap(BTRFS_I(inode), fieinfo, start, len); } -#define BTRFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC) - -static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, - __u64 start, __u64 len) +static int btrfs_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - int ret; - - ret = fiemap_check_flags(fieinfo, BTRFS_FIEMAP_FLAGS); - if (ret) - return ret; - - return extent_fiemap(inode, fieinfo, start, len); + return extent_writepages(mapping, wbc); } -int btrfs_readpage(struct file *file, struct page *page) +static void btrfs_readahead(struct readahead_control *rac) { - struct extent_io_tree *tree; - tree = &BTRFS_I(page->mapping->host)->io_tree; - return extent_read_full_page(tree, page, btrfs_get_extent, 0); + extent_readahead(rac); } -static int btrfs_writepage(struct page *page, struct writeback_control *wbc) +/* + * For release_folio() and invalidate_folio() we have a race window where + * folio_end_writeback() is called but the subpage spinlock is not yet released. + * If we continue to release/invalidate the page, we could cause use-after-free + * for subpage spinlock. So this function is to spin and wait for subpage + * spinlock. + */ +static void wait_subpage_spinlock(struct page *page) { - struct inode *inode = page->mapping->host; - int ret; + struct btrfs_fs_info *fs_info = btrfs_sb(page->mapping->host->i_sb); + struct btrfs_subpage *subpage; - if (current->flags & PF_MEMALLOC) { - redirty_page_for_writepage(wbc, page); - unlock_page(page); - return 0; - } + if (!btrfs_is_subpage(fs_info, page)) + return; + + ASSERT(PagePrivate(page) && page->private); + subpage = (struct btrfs_subpage *)page->private; /* - * If we are under memory pressure we will call this directly from the - * VM, we need to make sure we have the inode referenced for the ordered - * extent. If not just return like we didn't do anything. + * This may look insane as we just acquire the spinlock and release it, + * without doing anything. But we just want to make sure no one is + * still holding the subpage spinlock. + * And since the page is not dirty nor writeback, and we have page + * locked, the only possible way to hold a spinlock is from the endio + * function to clear page writeback. + * + * Here we just acquire the spinlock so that all existing callers + * should exit and we're safe to release/invalidate the page. */ - if (!igrab(inode)) { - redirty_page_for_writepage(wbc, page); - return AOP_WRITEPAGE_ACTIVATE; - } - ret = extent_write_full_page(page, wbc); - btrfs_add_delayed_iput(inode); - return ret; + spin_lock_irq(&subpage->lock); + spin_unlock_irq(&subpage->lock); } -static int btrfs_writepages(struct address_space *mapping, - struct writeback_control *wbc) +static bool __btrfs_release_folio(struct folio *folio, gfp_t gfp_flags) { - return extent_writepages(mapping, wbc); + int ret = try_release_extent_mapping(&folio->page, gfp_flags); + + if (ret == 1) { + wait_subpage_spinlock(&folio->page); + clear_page_extent_mapped(&folio->page); + } + return ret; } -static int -btrfs_readpages(struct file *file, struct address_space *mapping, - struct list_head *pages, unsigned nr_pages) +static bool btrfs_release_folio(struct folio *folio, gfp_t gfp_flags) { - return extent_readpages(mapping, pages, nr_pages); + if (folio_test_writeback(folio) || folio_test_dirty(folio)) + return false; + return __btrfs_release_folio(folio, gfp_flags); } -static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags) +#ifdef CONFIG_MIGRATION +static int btrfs_migrate_folio(struct address_space *mapping, + struct folio *dst, struct folio *src, + enum migrate_mode mode) { - int ret = try_release_extent_mapping(page, gfp_flags); - if (ret == 1) { - ClearPagePrivate(page); - set_page_private(page, 0); - put_page(page); + int ret = filemap_migrate_folio(mapping, dst, src, mode); + + if (ret != MIGRATEPAGE_SUCCESS) + return ret; + + if (folio_test_ordered(src)) { + folio_clear_ordered(src); + folio_set_ordered(dst); } - return ret; -} -static int btrfs_releasepage(struct page *page, gfp_t gfp_flags) -{ - if (PageWriteback(page) || PageDirty(page)) - return 0; - return __btrfs_releasepage(page, gfp_flags); + return MIGRATEPAGE_SUCCESS; } +#else +#define btrfs_migrate_folio NULL +#endif -static void btrfs_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) +static void btrfs_invalidate_folio(struct folio *folio, size_t offset, + size_t length) { - struct inode *inode = page->mapping->host; - struct extent_io_tree *tree; - struct btrfs_ordered_extent *ordered; + struct btrfs_inode *inode = BTRFS_I(folio->mapping->host); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_io_tree *tree = &inode->io_tree; struct extent_state *cached_state = NULL; - u64 page_start = page_offset(page); - u64 page_end = page_start + PAGE_SIZE - 1; - u64 start; - u64 end; - int inode_evicting = inode->i_state & I_FREEING; + u64 page_start = folio_pos(folio); + u64 page_end = page_start + folio_size(folio) - 1; + u64 cur; + int inode_evicting = inode->vfs_inode.i_state & I_FREEING; /* - * we have the page locked, so new writeback can't start, - * and the dirty bit won't be cleared while we are here. + * We have folio locked so no new ordered extent can be created on this + * page, nor bio can be submitted for this folio. + * + * But already submitted bio can still be finished on this folio. + * Furthermore, endio function won't skip folio which has Ordered + * (Private2) already cleared, so it's possible for endio and + * invalidate_folio to do the same ordered extent accounting twice + * on one folio. * - * Wait for IO on this page so that we can safely clear - * the PagePrivate2 bit and do ordered accounting + * So here we wait for any submitted bios to finish, so that we won't + * do double ordered extent accounting on the same folio. */ - wait_on_page_writeback(page); + folio_wait_writeback(folio); + wait_subpage_spinlock(&folio->page); - tree = &BTRFS_I(inode)->io_tree; - if (offset) { - btrfs_releasepage(page, GFP_NOFS); + /* + * For subpage case, we have call sites like + * btrfs_punch_hole_lock_range() which passes range not aligned to + * sectorsize. + * If the range doesn't cover the full folio, we don't need to and + * shouldn't clear page extent mapped, as folio->private can still + * record subpage dirty bits for other part of the range. + * + * For cases that invalidate the full folio even the range doesn't + * cover the full folio, like invalidating the last folio, we're + * still safe to wait for ordered extent to finish. + */ + if (!(offset == 0 && length == folio_size(folio))) { + btrfs_release_folio(folio, GFP_NOFS); return; } if (!inode_evicting) - lock_extent_bits(tree, page_start, page_end, &cached_state); -again: - start = page_start; - ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start, - page_end - start + 1); - if (ordered) { - end = min(page_end, - ordered->file_offset + ordered->num_bytes - 1); + lock_extent(tree, page_start, page_end, &cached_state); + + cur = page_start; + while (cur < page_end) { + struct btrfs_ordered_extent *ordered; + u64 range_end; + u32 range_len; + u32 extra_flags = 0; + + ordered = btrfs_lookup_first_ordered_range(inode, cur, + page_end + 1 - cur); + if (!ordered) { + range_end = page_end; + /* + * No ordered extent covering this range, we are safe + * to delete all extent states in the range. + */ + extra_flags = EXTENT_CLEAR_ALL_BITS; + goto next; + } + if (ordered->file_offset > cur) { + /* + * There is a range between [cur, oe->file_offset) not + * covered by any ordered extent. + * We are safe to delete all extent states, and handle + * the ordered extent in the next iteration. + */ + range_end = ordered->file_offset - 1; + extra_flags = EXTENT_CLEAR_ALL_BITS; + goto next; + } + + range_end = min(ordered->file_offset + ordered->num_bytes - 1, + page_end); + ASSERT(range_end + 1 - cur < U32_MAX); + range_len = range_end + 1 - cur; + if (!btrfs_page_test_ordered(fs_info, &folio->page, cur, range_len)) { + /* + * If Ordered (Private2) is cleared, it means endio has + * already been executed for the range. + * We can't delete the extent states as + * btrfs_finish_ordered_io() may still use some of them. + */ + goto next; + } + btrfs_page_clear_ordered(fs_info, &folio->page, cur, range_len); + /* - * IO on this page will never be started, so we need - * to account for any ordered extents now + * IO on this page will never be started, so we need to account + * for any ordered extents now. Don't clear EXTENT_DELALLOC_NEW + * here, must leave that up for the ordered extent completion. + * + * This will also unlock the range for incoming + * btrfs_finish_ordered_io(). */ if (!inode_evicting) - clear_extent_bit(tree, start, end, - EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | + clear_extent_bit(tree, cur, range_end, + EXTENT_DELALLOC | EXTENT_LOCKED | EXTENT_DO_ACCOUNTING | - EXTENT_DEFRAG, 1, 0, &cached_state); - /* - * whoever cleared the private bit is responsible - * for the finish_ordered_io - */ - if (TestClearPagePrivate2(page)) { - struct btrfs_ordered_inode_tree *tree; - u64 new_len; + EXTENT_DEFRAG, &cached_state); - tree = &BTRFS_I(inode)->ordered_tree; + spin_lock_irq(&inode->ordered_tree.lock); + set_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags); + ordered->truncated_len = min(ordered->truncated_len, + cur - ordered->file_offset); + spin_unlock_irq(&inode->ordered_tree.lock); - spin_lock_irq(&tree->lock); - set_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags); - new_len = start - ordered->file_offset; - if (new_len < ordered->truncated_len) - ordered->truncated_len = new_len; - spin_unlock_irq(&tree->lock); - - if (btrfs_dec_test_ordered_pending(inode, &ordered, - start, - end - start + 1, 1)) - btrfs_finish_ordered_io(ordered); + /* + * If the ordered extent has finished, we're safe to delete all + * the extent states of the range, otherwise + * btrfs_finish_ordered_io() will get executed by endio for + * other pages, so we can't delete extent states. + */ + if (btrfs_dec_test_ordered_pending(inode, &ordered, + cur, range_end + 1 - cur)) { + btrfs_finish_ordered_io(ordered); + /* + * The ordered extent has finished, now we're again + * safe to delete all extent states of the range. + */ + extra_flags = EXTENT_CLEAR_ALL_BITS; } - btrfs_put_ordered_extent(ordered); +next: + if (ordered) + btrfs_put_ordered_extent(ordered); + /* + * Qgroup reserved space handler + * Sector(s) here will be either: + * + * 1) Already written to disk or bio already finished + * Then its QGROUP_RESERVED bit in io_tree is already cleared. + * Qgroup will be handled by its qgroup_record then. + * btrfs_qgroup_free_data() call will do nothing here. + * + * 2) Not written to disk yet + * Then btrfs_qgroup_free_data() call will clear the + * QGROUP_RESERVED bit of its io_tree, and free the qgroup + * reserved data space. + * Since the IO will never happen for this page. + */ + btrfs_qgroup_free_data(inode, NULL, cur, range_end + 1 - cur); if (!inode_evicting) { - cached_state = NULL; - lock_extent_bits(tree, start, end, - &cached_state); + clear_extent_bit(tree, cur, range_end, EXTENT_LOCKED | + EXTENT_DELALLOC | EXTENT_UPTODATE | + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG | + extra_flags, &cached_state); } - - start = end + 1; - if (start < page_end) - goto again; + cur = range_end + 1; } - /* - * Qgroup reserved space handler - * Page here will be either - * 1) Already written to disk - * In this case, its reserved space is released from data rsv map - * and will be freed by delayed_ref handler finally. - * So even we call qgroup_free_data(), it won't decrease reserved - * space. - * 2) Not written to disk - * This means the reserved space should be freed here. However, - * if a truncate invalidates the page (by clearing PageDirty) - * and the page is accounted for while allocating extent - * in btrfs_check_data_free_space() we let delayed_ref to - * free the entire extent. - */ - if (PageDirty(page)) - btrfs_qgroup_free_data(inode, NULL, page_start, PAGE_SIZE); - if (!inode_evicting) { - clear_extent_bit(tree, page_start, page_end, EXTENT_LOCKED | - EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | - EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 1, 1, - &cached_state); - - __btrfs_releasepage(page, GFP_NOFS); - } - - ClearPageChecked(page); - if (PagePrivate(page)) { - ClearPagePrivate(page); - set_page_private(page, 0); - put_page(page); - } + * We have iterated through all ordered extents of the page, the page + * should not have Ordered (Private2) anymore, or the above iteration + * did something wrong. + */ + ASSERT(!folio_test_ordered(folio)); + btrfs_page_clear_checked(fs_info, &folio->page, folio_pos(folio), folio_size(folio)); + if (!inode_evicting) + __btrfs_release_folio(folio, GFP_NOFS); + clear_page_extent_mapped(&folio->page); } /* @@ -8410,7 +8400,6 @@ vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; struct extent_changeset *data_reserved = NULL; - char *kaddr; unsigned long zero_start; loff_t size; vm_fault_t ret; @@ -8432,12 +8421,12 @@ vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) * Reserving delalloc space after obtaining the page lock can lead to * deadlock. For example, if a dirty page is locked by this function * and the call to btrfs_delalloc_reserve_space() ends up triggering - * dirty page write out, then the btrfs_writepage() function could + * dirty page write out, then the btrfs_writepages() function could * end up waiting indefinitely to get a lock on the page currently * being processed by btrfs_page_mkwrite() function. */ - ret2 = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start, - reserved_space); + ret2 = btrfs_delalloc_reserve_space(BTRFS_I(inode), &data_reserved, + page_start, reserved_space); if (!ret2) { ret2 = file_update_time(vmf->vma->vm_file); reserved = 1; @@ -8451,6 +8440,7 @@ vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */ again: + down_read(&BTRFS_I(inode)->i_mmap_lock); lock_page(page); size = i_size_read(inode); @@ -8461,8 +8451,13 @@ again: } wait_on_page_writeback(page); - lock_extent_bits(io_tree, page_start, page_end, &cached_state); - set_page_extent_mapped(page); + lock_extent(io_tree, page_start, page_end, &cached_state); + ret2 = set_page_extent_mapped(page); + if (ret2 < 0) { + ret = vmf_error(ret2); + unlock_extent(io_tree, page_start, page_end, &cached_state); + goto out_unlock; + } /* * we can't set the delalloc bits if there are pending ordered @@ -8471,10 +8466,10 @@ again: 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); + unlock_extent(io_tree, page_start, page_end, &cached_state); unlock_page(page); - btrfs_start_ordered_extent(inode, ordered, 1); + up_read(&BTRFS_I(inode)->i_mmap_lock); + btrfs_start_ordered_extent(ordered, 1); btrfs_put_ordered_extent(ordered); goto again; } @@ -8484,9 +8479,9 @@ again: fs_info->sectorsize); if (reserved_space < PAGE_SIZE) { end = page_start + reserved_space - 1; - btrfs_delalloc_release_space(inode, data_reserved, - page_start, PAGE_SIZE - reserved_space, - true); + btrfs_delalloc_release_space(BTRFS_I(inode), + data_reserved, page_start, + PAGE_SIZE - reserved_space, true); } } @@ -8499,13 +8494,12 @@ again: */ clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end, EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | - EXTENT_DEFRAG, 0, 0, &cached_state); + EXTENT_DEFRAG, &cached_state); - ret2 = btrfs_set_extent_delalloc(inode, page_start, end, 0, + ret2 = btrfs_set_extent_delalloc(BTRFS_I(inode), page_start, end, 0, &cached_state); if (ret2) { - unlock_extent_cached(io_tree, page_start, page_end, - &cached_state); + unlock_extent(io_tree, page_start, page_end, &cached_state); ret = VM_FAULT_SIGBUS; goto out_unlock; } @@ -8516,21 +8510,17 @@ again: else zero_start = PAGE_SIZE; - if (zero_start != PAGE_SIZE) { - kaddr = kmap(page); - memset(kaddr + zero_start, 0, PAGE_SIZE - zero_start); - flush_dcache_page(page); - kunmap(page); - } - ClearPageChecked(page); - set_page_dirty(page); - SetPageUptodate(page); + if (zero_start != PAGE_SIZE) + memzero_page(page, zero_start, PAGE_SIZE - zero_start); - 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; + btrfs_page_clear_checked(fs_info, page, page_start, PAGE_SIZE); + btrfs_page_set_dirty(fs_info, page, page_start, end + 1 - page_start); + btrfs_page_set_uptodate(fs_info, page, page_start, end + 1 - page_start); - unlock_extent_cached(io_tree, page_start, page_end, &cached_state); + btrfs_set_inode_last_sub_trans(BTRFS_I(inode)); + + unlock_extent(io_tree, page_start, page_end, &cached_state); + up_read(&BTRFS_I(inode)->i_mmap_lock); btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); sb_end_pagefault(inode->i_sb); @@ -8539,9 +8529,10 @@ again: out_unlock: unlock_page(page); + up_read(&BTRFS_I(inode)->i_mmap_lock); out: btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); - btrfs_delalloc_release_space(inode, data_reserved, page_start, + btrfs_delalloc_release_space(BTRFS_I(inode), data_reserved, page_start, reserved_space, (ret != 0)); out_noreserve: sb_end_pagefault(inode->i_sb); @@ -8551,6 +8542,12 @@ out_noreserve: static int btrfs_truncate(struct inode *inode, bool skip_writeback) { + struct btrfs_truncate_control control = { + .inode = BTRFS_I(inode), + .ino = btrfs_ino(BTRFS_I(inode)), + .min_type = BTRFS_EXTENT_DATA_KEY, + .clear_extent_range = true, + }; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *rsv; @@ -8598,7 +8595,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) if (!rsv) return -ENOMEM; rsv->size = min_size; - rsv->failfast = 1; + rsv->failfast = true; /* * 1 for the truncate slack space @@ -8615,25 +8612,38 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) min_size, false); BUG_ON(ret); - /* - * So if we truncate and then write and fsync we normally would just - * write the extents that changed, which is a problem if we need to - * first truncate that entire inode. So set this flag so we write out - * all of the extents in the inode to the sync log so we're completely - * safe. - */ - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); trans->block_rsv = rsv; while (1) { - ret = btrfs_truncate_inode_items(trans, root, inode, - inode->i_size, - BTRFS_EXTENT_DATA_KEY); + struct extent_state *cached_state = NULL; + const u64 new_size = inode->i_size; + const u64 lock_start = ALIGN_DOWN(new_size, fs_info->sectorsize); + + control.new_size = new_size; + lock_extent(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, + &cached_state); + /* + * We want to drop from the next block forward in case this new + * size is not block aligned since we will be keeping the last + * block of the extent just the way it is. + */ + btrfs_drop_extent_map_range(BTRFS_I(inode), + ALIGN(new_size, fs_info->sectorsize), + (u64)-1, false); + + ret = btrfs_truncate_inode_items(trans, root, &control); + + inode_sub_bytes(inode, control.sub_bytes); + btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), control.last_size); + + unlock_extent(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, + &cached_state); + trans->block_rsv = &fs_info->trans_block_rsv; if (ret != -ENOSPC && ret != -EAGAIN) break; - ret = btrfs_update_inode(trans, root, inode); + ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); if (ret) break; @@ -8647,7 +8657,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) break; } - btrfs_block_rsv_release(fs_info, rsv, -1); + btrfs_block_rsv_release(fs_info, rsv, -1, NULL); ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, false); BUG_ON(ret); /* shouldn't happen */ @@ -8656,15 +8666,15 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) /* * We can't call btrfs_truncate_block inside a trans handle as we could - * deadlock with freeze, if we got NEED_TRUNCATE_BLOCK then we know - * we've truncated everything except the last little bit, and can do - * btrfs_truncate_block and then update the disk_i_size. + * deadlock with freeze, if we got BTRFS_NEED_TRUNCATE_BLOCK then we + * know we've truncated everything except the last little bit, and can + * do btrfs_truncate_block and then update the disk_i_size. */ - if (ret == NEED_TRUNCATE_BLOCK) { + if (ret == BTRFS_NEED_TRUNCATE_BLOCK) { btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); - ret = btrfs_truncate_block(inode, inode->i_size, 0, 0); + ret = btrfs_truncate_block(BTRFS_I(inode), inode->i_size, 0, 0); if (ret) goto out; trans = btrfs_start_transaction(root, 1); @@ -8672,14 +8682,14 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) ret = PTR_ERR(trans); goto out; } - btrfs_ordered_update_i_size(inode, inode->i_size, NULL); + btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0); } if (trans) { int ret2; trans->block_rsv = &fs_info->trans_block_rsv; - ret2 = btrfs_update_inode(trans, root, inode); + ret2 = btrfs_update_inode(trans, root, BTRFS_I(inode)); if (ret2 && !ret) ret = ret2; @@ -8690,45 +8700,43 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) } out: btrfs_free_block_rsv(fs_info, rsv); + /* + * So if we truncate and then write and fsync we normally would just + * write the extents that changed, which is a problem if we need to + * first truncate that entire inode. So set this flag so we write out + * all of the extents in the inode to the sync log so we're completely + * safe. + * + * If no extents were dropped or trimmed we don't need to force the next + * fsync to truncate all the inode's items from the log and re-log them + * all. This means the truncate operation did not change the file size, + * or changed it to a smaller size but there was only an implicit hole + * between the old i_size and the new i_size, and there were no prealloc + * extents beyond i_size to drop. + */ + if (control.extents_found > 0) + btrfs_set_inode_full_sync(BTRFS_I(inode)); return ret; } -/* - * create a new subvolume directory/inode (helper for the ioctl). - */ -int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, - struct btrfs_root *new_root, - struct btrfs_root *parent_root, - u64 new_dirid) +struct inode *btrfs_new_subvol_inode(struct user_namespace *mnt_userns, + struct inode *dir) { struct inode *inode; - int err; - u64 index = 0; - - inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, - new_dirid, new_dirid, - S_IFDIR | (~current_umask() & S_IRWXUGO), - &index); - if (IS_ERR(inode)) - return PTR_ERR(inode); - inode->i_op = &btrfs_dir_inode_operations; - inode->i_fop = &btrfs_dir_file_operations; - - set_nlink(inode, 1); - btrfs_i_size_write(BTRFS_I(inode), 0); - unlock_new_inode(inode); - err = btrfs_subvol_inherit_props(trans, new_root, parent_root); - if (err) - btrfs_err(new_root->fs_info, - "error inheriting subvolume %llu properties: %d", - new_root->root_key.objectid, err); - - err = btrfs_update_inode(trans, new_root, inode); - - iput(inode); - return err; + inode = new_inode(dir->i_sb); + if (inode) { + /* + * Subvolumes don't inherit the sgid bit or the parent's gid if + * the parent's sgid bit is set. This is probably a bug. + */ + inode_init_owner(mnt_userns, inode, NULL, + S_IFDIR | (~current_umask() & S_IRWXUGO)); + inode->i_op = &btrfs_dir_inode_operations; + inode->i_fop = &btrfs_dir_file_operations; + } + return inode; } struct inode *btrfs_alloc_inode(struct super_block *sb) @@ -8737,7 +8745,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) struct btrfs_inode *ei; struct inode *inode; - ei = kmem_cache_alloc(btrfs_inode_cachep, GFP_KERNEL); + ei = alloc_inode_sb(sb, btrfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; @@ -8751,13 +8759,16 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->defrag_bytes = 0; ei->disk_i_size = 0; ei->flags = 0; + ei->ro_flags = 0; ei->csum_bytes = 0; ei->index_cnt = (u64)-1; ei->dir_index = 0; ei->last_unlink_trans = 0; + ei->last_reflink_trans = 0; ei->last_log_commit = 0; spin_lock_init(&ei->lock); + spin_lock_init(&ei->io_failure_lock); ei->outstanding_extents = 0; if (sb->s_magic != BTRFS_TEST_MAGIC) btrfs_init_metadata_block_rsv(fs_info, &ei->block_rsv, @@ -8774,17 +8785,16 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) inode = &ei->vfs_inode; extent_map_tree_init(&ei->extent_tree); extent_io_tree_init(fs_info, &ei->io_tree, IO_TREE_INODE_IO, inode); - extent_io_tree_init(fs_info, &ei->io_failure_tree, - IO_TREE_INODE_IO_FAILURE, inode); - ei->io_tree.track_uptodate = true; - ei->io_failure_tree.track_uptodate = true; + extent_io_tree_init(fs_info, &ei->file_extent_tree, + IO_TREE_INODE_FILE_EXTENT, NULL); + ei->io_failure_tree = RB_ROOT; atomic_set(&ei->sync_writers, 0); mutex_init(&ei->log_mutex); btrfs_ordered_inode_tree_init(&ei->ordered_tree); INIT_LIST_HEAD(&ei->delalloc_inodes); INIT_LIST_HEAD(&ei->delayed_iput); RB_CLEAR_NODE(&ei->rb_node); - init_rwsem(&ei->dio_sem); + init_rwsem(&ei->i_mmap_lock); return inode; } @@ -8792,7 +8802,7 @@ 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(BTRFS_I(inode), 0, (u64)-1, 0); + btrfs_drop_extent_map_range(BTRFS_I(inode), 0, (u64)-1, false); kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); } #endif @@ -8802,21 +8812,24 @@ void btrfs_free_inode(struct inode *inode) kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); } -void btrfs_destroy_inode(struct inode *inode) +void btrfs_destroy_inode(struct inode *vfs_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; + struct btrfs_inode *inode = BTRFS_I(vfs_inode); + struct btrfs_root *root = inode->root; + bool freespace_inode; - WARN_ON(!hlist_empty(&inode->i_dentry)); - WARN_ON(inode->i_data.nrpages); - WARN_ON(BTRFS_I(inode)->block_rsv.reserved); - WARN_ON(BTRFS_I(inode)->block_rsv.size); - WARN_ON(BTRFS_I(inode)->outstanding_extents); - WARN_ON(BTRFS_I(inode)->delalloc_bytes); - WARN_ON(BTRFS_I(inode)->new_delalloc_bytes); - WARN_ON(BTRFS_I(inode)->csum_bytes); - WARN_ON(BTRFS_I(inode)->defrag_bytes); + WARN_ON(!hlist_empty(&vfs_inode->i_dentry)); + WARN_ON(vfs_inode->i_data.nrpages); + WARN_ON(inode->block_rsv.reserved); + WARN_ON(inode->block_rsv.size); + WARN_ON(inode->outstanding_extents); + if (!S_ISDIR(vfs_inode->i_mode)) { + WARN_ON(inode->delalloc_bytes); + WARN_ON(inode->new_delalloc_bytes); + } + WARN_ON(inode->csum_bytes); + WARN_ON(inode->defrag_bytes); /* * This can happen where we create an inode, but somebody else also @@ -8826,14 +8839,24 @@ void btrfs_destroy_inode(struct inode *inode) if (!root) return; + /* + * If this is a free space inode do not take the ordered extents lockdep + * map. + */ + freespace_inode = btrfs_is_free_space_inode(inode); + while (1) { ordered = btrfs_lookup_first_ordered_extent(inode, (u64)-1); if (!ordered) break; else { - btrfs_err(fs_info, + btrfs_err(root->fs_info, "found ordered extent %llu %llu on inode cleanup", ordered->file_offset, ordered->num_bytes); + + if (!freespace_inode) + btrfs_lockdep_acquire(root->fs_info, btrfs_ordered_extent); + btrfs_remove_ordered_extent(inode, ordered); btrfs_put_ordered_extent(ordered); btrfs_put_ordered_extent(ordered); @@ -8841,7 +8864,9 @@ void btrfs_destroy_inode(struct inode *inode) } btrfs_qgroup_check_reserved_leak(inode); inode_tree_del(inode); - btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0); + btrfs_drop_extent_map_range(inode, 0, (u64)-1, false); + btrfs_inode_clear_file_extent_range(inode, 0, (u64)-1); + btrfs_put_root(inode->root); } int btrfs_drop_inode(struct inode *inode) @@ -8860,7 +8885,7 @@ int btrfs_drop_inode(struct inode *inode) static void init_once(void *foo) { - struct btrfs_inode *ei = (struct btrfs_inode *) foo; + struct btrfs_inode *ei = foo; inode_init_once(&ei->vfs_inode); } @@ -8872,6 +8897,7 @@ void __cold btrfs_destroy_cachep(void) * destroy cache. */ rcu_barrier(); + bioset_exit(&btrfs_dio_bioset); kmem_cache_destroy(btrfs_inode_cachep); kmem_cache_destroy(btrfs_trans_handle_cachep); kmem_cache_destroy(btrfs_path_cachep); @@ -8908,23 +8934,31 @@ int __init btrfs_init_cachep(void) btrfs_free_space_bitmap_cachep = kmem_cache_create("btrfs_free_space_bitmap", PAGE_SIZE, PAGE_SIZE, - SLAB_RED_ZONE, NULL); + SLAB_MEM_SPREAD, NULL); if (!btrfs_free_space_bitmap_cachep) goto fail; + if (bioset_init(&btrfs_dio_bioset, BIO_POOL_SIZE, + offsetof(struct btrfs_dio_private, bio), + BIOSET_NEED_BVECS)) + goto fail; + return 0; fail: btrfs_destroy_cachep(); return -ENOMEM; } -static int btrfs_getattr(const struct path *path, struct kstat *stat, +static int btrfs_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { u64 delalloc_bytes; + u64 inode_bytes; struct inode *inode = d_inode(path->dentry); u32 blocksize = inode->i_sb->s_blocksize; u32 bi_flags = BTRFS_I(inode)->flags; + u32 bi_ro_flags = BTRFS_I(inode)->ro_flags; stat->result_mask |= STATX_BTIME; stat->btime.tv_sec = BTRFS_I(inode)->i_otime.tv_sec; @@ -8937,19 +8971,22 @@ static int btrfs_getattr(const struct path *path, struct kstat *stat, stat->attributes |= STATX_ATTR_IMMUTABLE; if (bi_flags & BTRFS_INODE_NODUMP) stat->attributes |= STATX_ATTR_NODUMP; + if (bi_ro_flags & BTRFS_INODE_RO_VERITY) + stat->attributes |= STATX_ATTR_VERITY; stat->attributes_mask |= (STATX_ATTR_APPEND | STATX_ATTR_COMPRESSED | STATX_ATTR_IMMUTABLE | STATX_ATTR_NODUMP); - generic_fillattr(inode, stat); + generic_fillattr(mnt_userns, inode, stat); stat->dev = BTRFS_I(inode)->root->anon_dev; spin_lock(&BTRFS_I(inode)->lock); delalloc_bytes = BTRFS_I(inode)->new_delalloc_bytes; + inode_bytes = inode_get_bytes(inode); spin_unlock(&BTRFS_I(inode)->lock); - stat->blocks = (ALIGN(inode_get_bytes(inode), blocksize) + + stat->blocks = (ALIGN(inode_bytes, blocksize) + ALIGN(delalloc_bytes, blocksize)) >> 9; return 0; } @@ -8961,53 +8998,79 @@ static int btrfs_rename_exchange(struct inode *old_dir, { 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; struct btrfs_root *dest = BTRFS_I(new_dir)->root; struct inode *new_inode = new_dentry->d_inode; struct inode *old_inode = old_dentry->d_inode; struct timespec64 ctime = current_time(old_inode); - struct dentry *parent; + struct btrfs_rename_ctx old_rename_ctx; + struct btrfs_rename_ctx new_rename_ctx; 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; int ret; - bool root_log_pinned = false; - bool dest_log_pinned = false; - struct btrfs_log_ctx ctx_root; - struct btrfs_log_ctx ctx_dest; - bool sync_log_root = false; - bool sync_log_dest = false; - bool commit_transaction = false; + int ret2; + bool need_abort = false; - /* we only allow rename subvolume link between subvolumes */ - if (old_ino != BTRFS_FIRST_FREE_OBJECTID && root != dest) + /* + * For non-subvolumes allow exchange only within one subvolume, in the + * same inode namespace. Two subvolumes (represented as directory) can + * be exchanged as they're a logical link and have a fixed inode number. + */ + if (root != dest && + (old_ino != BTRFS_FIRST_FREE_OBJECTID || + new_ino != BTRFS_FIRST_FREE_OBJECTID)) return -EXDEV; - btrfs_init_log_ctx(&ctx_root, old_inode); - btrfs_init_log_ctx(&ctx_dest, new_inode); - /* close the race window with snapshot create/destroy ioctl */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID || new_ino == BTRFS_FIRST_FREE_OBJECTID) 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 - * require 4 item modifications, but if they are both normal inodes it - * would require 5 item modifications, so we'll assume their normal - * inodes. So 5 * 2 is 10, plus 2 for the new links, so 12 total items - * should cover the worst case number of items we'll modify. + * For each inode: + * 1 to remove old dir item + * 1 to remove old dir index + * 1 to add new dir item + * 1 to add new dir index + * 1 to update parent inode + * + * If the parents are the same, we only need to account for one */ - trans = btrfs_start_transaction(root, 12); + trans_num_items = (old_dir == new_dir ? 9 : 10); + if (old_ino == BTRFS_FIRST_FREE_OBJECTID) { + /* + * 1 to remove old root ref + * 1 to remove old root backref + * 1 to add new root ref + * 1 to add new root backref + */ + trans_num_items += 4; + } else { + /* + * 1 to update inode item + * 1 to remove old inode ref + * 1 to add new inode ref + */ + trans_num_items += 3; + } + if (new_ino == BTRFS_FIRST_FREE_OBJECTID) + trans_num_items += 4; + else + trans_num_items += 3; + trans = btrfs_start_transaction(root, trans_num_items); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out_notrans; } - if (dest != root) - btrfs_record_root_in_trans(trans, dest); + if (dest != root) { + ret = btrfs_record_root_in_trans(trans, dest); + if (ret) + goto out_fail; + } /* * We need to find a free sequence number both in the source and @@ -9028,8 +9091,6 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* force full log commit if subvolume involved. */ btrfs_set_log_full_commit(trans); } else { - btrfs_pin_log_trans(root); - root_log_pinned = true; ret = btrfs_insert_inode_ref(trans, dest, new_dentry->d_name.name, new_dentry->d_name.len, @@ -9038,6 +9099,7 @@ static int btrfs_rename_exchange(struct inode *old_dir, old_idx); if (ret) goto out_fail; + need_abort = true; } /* And now for the dest. */ @@ -9045,16 +9107,17 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* force full log commit if subvolume involved. */ btrfs_set_log_full_commit(trans); } else { - btrfs_pin_log_trans(dest); - dest_log_pinned = true; ret = btrfs_insert_inode_ref(trans, root, old_dentry->d_name.name, old_dentry->d_name.len, new_ino, btrfs_ino(BTRFS_I(old_dir)), new_idx); - if (ret) + if (ret) { + if (need_abort) + btrfs_abort_transaction(trans, ret); goto out_fail; + } } /* Update inode version and ctime/mtime. */ @@ -9062,8 +9125,10 @@ static int btrfs_rename_exchange(struct inode *old_dir, inode_inc_iversion(new_dir); inode_inc_iversion(old_inode); inode_inc_iversion(new_inode); - old_dir->i_ctime = old_dir->i_mtime = ctime; - new_dir->i_ctime = new_dir->i_mtime = ctime; + old_dir->i_mtime = ctime; + old_dir->i_ctime = ctime; + new_dir->i_mtime = ctime; + new_dir->i_ctime = ctime; old_inode->i_ctime = ctime; new_inode->i_ctime = ctime; @@ -9078,12 +9143,13 @@ static int btrfs_rename_exchange(struct inode *old_dir, if (old_ino == BTRFS_FIRST_FREE_OBJECTID) { ret = btrfs_unlink_subvol(trans, old_dir, old_dentry); } else { /* src is an inode */ - ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir), + ret = __btrfs_unlink_inode(trans, BTRFS_I(old_dir), BTRFS_I(old_dentry->d_inode), old_dentry->d_name.name, - old_dentry->d_name.len); + old_dentry->d_name.len, + &old_rename_ctx); if (!ret) - ret = btrfs_update_inode(trans, root, old_inode); + ret = btrfs_update_inode(trans, root, BTRFS_I(old_inode)); } if (ret) { btrfs_abort_transaction(trans, ret); @@ -9094,12 +9160,13 @@ static int btrfs_rename_exchange(struct inode *old_dir, if (new_ino == BTRFS_FIRST_FREE_OBJECTID) { ret = btrfs_unlink_subvol(trans, new_dir, new_dentry); } else { /* dest is an inode */ - ret = __btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir), + ret = __btrfs_unlink_inode(trans, BTRFS_I(new_dir), BTRFS_I(new_dentry->d_inode), new_dentry->d_name.name, - new_dentry->d_name.len); + new_dentry->d_name.len, + &new_rename_ctx); if (!ret) - ret = btrfs_update_inode(trans, dest, new_inode); + ret = btrfs_update_inode(trans, dest, BTRFS_I(new_inode)); } if (ret) { btrfs_abort_transaction(trans, ret); @@ -9127,175 +9194,77 @@ static int btrfs_rename_exchange(struct inode *old_dir, if (new_inode->i_nlink == 1) BTRFS_I(new_inode)->dir_index = new_idx; - if (root_log_pinned) { - parent = new_dentry->d_parent; - ret = btrfs_log_new_name(trans, BTRFS_I(old_inode), - BTRFS_I(old_dir), parent, - false, &ctx_root); - if (ret == BTRFS_NEED_LOG_SYNC) - sync_log_root = true; - else if (ret == BTRFS_NEED_TRANS_COMMIT) - commit_transaction = true; - ret = 0; + /* + * Now pin the logs of the roots. We do it to ensure that no other task + * can sync the logs while we are in progress with the rename, because + * that could result in an inconsistency in case any of the inodes that + * are part of this rename operation were logged before. + */ + if (old_ino != BTRFS_FIRST_FREE_OBJECTID) + btrfs_pin_log_trans(root); + if (new_ino != BTRFS_FIRST_FREE_OBJECTID) + btrfs_pin_log_trans(dest); + + /* Do the log updates for all inodes. */ + if (old_ino != BTRFS_FIRST_FREE_OBJECTID) + btrfs_log_new_name(trans, old_dentry, BTRFS_I(old_dir), + old_rename_ctx.index, new_dentry->d_parent); + if (new_ino != BTRFS_FIRST_FREE_OBJECTID) + btrfs_log_new_name(trans, new_dentry, BTRFS_I(new_dir), + new_rename_ctx.index, old_dentry->d_parent); + + /* Now unpin the logs. */ + if (old_ino != BTRFS_FIRST_FREE_OBJECTID) btrfs_end_log_trans(root); - root_log_pinned = false; - } - if (dest_log_pinned) { - if (!commit_transaction) { - parent = old_dentry->d_parent; - ret = btrfs_log_new_name(trans, BTRFS_I(new_inode), - BTRFS_I(new_dir), parent, - false, &ctx_dest); - if (ret == BTRFS_NEED_LOG_SYNC) - sync_log_dest = true; - else if (ret == BTRFS_NEED_TRANS_COMMIT) - commit_transaction = true; - ret = 0; - } + if (new_ino != BTRFS_FIRST_FREE_OBJECTID) btrfs_end_log_trans(dest); - dest_log_pinned = false; - } out_fail: - /* - * If we have pinned a log and an error happened, we unpin tasks - * trying to sync the log and force them to fallback to a transaction - * commit if the log currently contains any of the inodes involved in - * this rename operation (to ensure we do not persist a log with an - * inconsistent state for any of these inodes or leading to any - * inconsistencies when replayed). If the transaction was aborted, the - * abortion reason is propagated to userspace when attempting to commit - * the transaction. If the log does not contain any of these inodes, we - * allow the tasks to sync it. - */ - if (ret && (root_log_pinned || dest_log_pinned)) { - 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(BTRFS_I(new_inode), fs_info->generation))) - btrfs_set_log_full_commit(trans); - - if (root_log_pinned) { - btrfs_end_log_trans(root); - root_log_pinned = false; - } - if (dest_log_pinned) { - btrfs_end_log_trans(dest); - dest_log_pinned = false; - } - } - if (!ret && sync_log_root && !commit_transaction) { - ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, - &ctx_root); - if (ret) - commit_transaction = true; - } - if (!ret && sync_log_dest && !commit_transaction) { - ret = btrfs_sync_log(trans, BTRFS_I(new_inode)->root, - &ctx_dest); - if (ret) - commit_transaction = true; - } - if (commit_transaction) { - /* - * We may have set commit_transaction when logging the new name - * in the destination root, in which case we left the source - * root context in the list of log contextes. So make sure we - * remove it to avoid invalid memory accesses, since the context - * was allocated in our stack frame. - */ - if (sync_log_root) { - mutex_lock(&root->log_mutex); - list_del_init(&ctx_root.list); - mutex_unlock(&root->log_mutex); - } - ret = btrfs_commit_transaction(trans); - } else { - int ret2; - - ret2 = btrfs_end_transaction(trans); - ret = ret ? ret : ret2; - } + ret2 = btrfs_end_transaction(trans); + ret = ret ? ret : ret2; out_notrans: if (new_ino == BTRFS_FIRST_FREE_OBJECTID || old_ino == BTRFS_FIRST_FREE_OBJECTID) up_read(&fs_info->subvol_sem); - ASSERT(list_empty(&ctx_root.list)); - ASSERT(list_empty(&ctx_dest.list)); - return ret; } -static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct inode *dir, - struct dentry *dentry) +static struct inode *new_whiteout_inode(struct user_namespace *mnt_userns, + struct inode *dir) { - int ret; struct inode *inode; - u64 objectid; - u64 index; - - ret = btrfs_find_free_ino(root, &objectid); - if (ret) - return ret; - - inode = btrfs_new_inode(trans, root, dir, - dentry->d_name.name, - dentry->d_name.len, - btrfs_ino(BTRFS_I(dir)), - objectid, - S_IFCHR | WHITEOUT_MODE, - &index); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - return ret; + inode = new_inode(dir->i_sb); + if (inode) { + inode_init_owner(mnt_userns, inode, dir, + S_IFCHR | WHITEOUT_MODE); + inode->i_op = &btrfs_special_inode_operations; + init_special_inode(inode, inode->i_mode, WHITEOUT_DEV); } - - inode->i_op = &btrfs_special_inode_operations; - init_special_inode(inode, inode->i_mode, - WHITEOUT_DEV); - - ret = btrfs_init_inode_security(trans, inode, dir, - &dentry->d_name); - if (ret) - goto out; - - ret = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, - BTRFS_I(inode), 0, index); - if (ret) - goto out; - - ret = btrfs_update_inode(trans, root, inode); -out: - unlock_new_inode(inode); - if (ret) - inode_dec_link_count(inode); - iput(inode); - - return ret; + return inode; } -static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int btrfs_rename(struct user_namespace *mnt_userns, + 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_new_inode_args whiteout_args = { + .dir = old_dir, + .dentry = old_dentry, + }; struct btrfs_trans_handle *trans; unsigned int trans_num_items; struct btrfs_root *root = BTRFS_I(old_dir)->root; struct btrfs_root *dest = BTRFS_I(new_dir)->root; struct inode *new_inode = d_inode(new_dentry); struct inode *old_inode = d_inode(old_dentry); + struct btrfs_rename_ctx rename_ctx; u64 index = 0; int ret; + int ret2; u64 old_ino = btrfs_ino(BTRFS_I(old_inode)); - bool log_pinned = false; - struct btrfs_log_ctx ctx; - bool sync_log = false; - bool commit_transaction = false; if (btrfs_ino(BTRFS_I(new_dir)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) return -EPERM; @@ -9339,31 +9308,67 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size) filemap_flush(old_inode->i_mapping); - /* close the racy window with snapshot create/destroy ioctl */ - if (old_ino == BTRFS_FIRST_FREE_OBJECTID) + if (flags & RENAME_WHITEOUT) { + whiteout_args.inode = new_whiteout_inode(mnt_userns, old_dir); + if (!whiteout_args.inode) + return -ENOMEM; + ret = btrfs_new_inode_prepare(&whiteout_args, &trans_num_items); + if (ret) + goto out_whiteout_inode; + } else { + /* 1 to update the old parent inode. */ + trans_num_items = 1; + } + + if (old_ino == BTRFS_FIRST_FREE_OBJECTID) { + /* Close the race window with snapshot create/destroy ioctl */ down_read(&fs_info->subvol_sem); + /* + * 1 to remove old root ref + * 1 to remove old root backref + * 1 to add new root ref + * 1 to add new root backref + */ + trans_num_items += 4; + } else { + /* + * 1 to update inode + * 1 to remove old inode ref + * 1 to add new inode ref + */ + trans_num_items += 3; + } /* - * 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 - * require 4 item modifications, but if they are both normal inodes it - * would require 5 item modifications, so we'll assume they are normal - * inodes. So 5 * 2 is 10, plus 1 for the new link, so 11 total items - * should cover the worst case number of items we'll modify. - * If our rename has the whiteout flag, we need more 5 units for the - * new inode (1 inode item, 1 inode ref, 2 dir items and 1 xattr item - * when selinux is enabled). - */ - trans_num_items = 11; - if (flags & RENAME_WHITEOUT) + * 1 to remove old dir item + * 1 to remove old dir index + * 1 to add new dir item + * 1 to add new dir index + */ + trans_num_items += 4; + /* 1 to update new parent inode if it's not the same as the old parent */ + if (new_dir != old_dir) + trans_num_items++; + if (new_inode) { + /* + * 1 to update inode + * 1 to remove inode ref + * 1 to remove dir item + * 1 to remove dir index + * 1 to possibly add orphan item + */ trans_num_items += 5; + } trans = btrfs_start_transaction(root, trans_num_items); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out_notrans; } - if (dest != root) - btrfs_record_root_in_trans(trans, dest); + if (dest != root) { + ret = btrfs_record_root_in_trans(trans, dest); + if (ret) + goto out_fail; + } ret = btrfs_set_inode_index(BTRFS_I(new_dir), &index); if (ret) @@ -9374,8 +9379,6 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, /* force full log commit if subvolume involved. */ btrfs_set_log_full_commit(trans); } else { - btrfs_pin_log_trans(root); - log_pinned = true; ret = btrfs_insert_inode_ref(trans, dest, new_dentry->d_name.name, new_dentry->d_name.len, @@ -9388,9 +9391,11 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, inode_inc_iversion(old_dir); inode_inc_iversion(new_dir); inode_inc_iversion(old_inode); - old_dir->i_ctime = old_dir->i_mtime = - new_dir->i_ctime = new_dir->i_mtime = - old_inode->i_ctime = current_time(old_dir); + old_dir->i_mtime = current_time(old_dir); + old_dir->i_ctime = old_dir->i_mtime; + new_dir->i_mtime = old_dir->i_mtime; + new_dir->i_ctime = old_dir->i_mtime; + old_inode->i_ctime = old_dir->i_mtime; if (old_dentry->d_parent != new_dentry->d_parent) btrfs_record_unlink_dir(trans, BTRFS_I(old_dir), @@ -9399,12 +9404,13 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) { ret = btrfs_unlink_subvol(trans, old_dir, old_dentry); } else { - ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir), + ret = __btrfs_unlink_inode(trans, BTRFS_I(old_dir), BTRFS_I(d_inode(old_dentry)), old_dentry->d_name.name, - old_dentry->d_name.len); + old_dentry->d_name.len, + &rename_ctx); if (!ret) - ret = btrfs_update_inode(trans, root, old_inode); + ret = btrfs_update_inode(trans, root, BTRFS_I(old_inode)); } if (ret) { btrfs_abort_transaction(trans, ret); @@ -9419,7 +9425,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, ret = btrfs_unlink_subvol(trans, new_dir, new_dentry); BUG_ON(new_inode->i_nlink == 0); } else { - ret = btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir), + ret = btrfs_unlink_inode(trans, BTRFS_I(new_dir), BTRFS_I(d_inode(new_dentry)), new_dentry->d_name.name, new_dentry->d_name.len); @@ -9444,90 +9450,54 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (old_inode->i_nlink == 1) BTRFS_I(old_inode)->dir_index = index; - if (log_pinned) { - struct dentry *parent = new_dentry->d_parent; - - btrfs_init_log_ctx(&ctx, old_inode); - ret = btrfs_log_new_name(trans, BTRFS_I(old_inode), - BTRFS_I(old_dir), parent, - false, &ctx); - if (ret == BTRFS_NEED_LOG_SYNC) - sync_log = true; - else if (ret == BTRFS_NEED_TRANS_COMMIT) - commit_transaction = true; - ret = 0; - btrfs_end_log_trans(root); - log_pinned = false; - } + if (old_ino != BTRFS_FIRST_FREE_OBJECTID) + btrfs_log_new_name(trans, old_dentry, BTRFS_I(old_dir), + rename_ctx.index, new_dentry->d_parent); if (flags & RENAME_WHITEOUT) { - ret = btrfs_whiteout_for_rename(trans, root, old_dir, - old_dentry); - + ret = btrfs_create_new_inode(trans, &whiteout_args); if (ret) { btrfs_abort_transaction(trans, ret); goto out_fail; + } else { + unlock_new_inode(whiteout_args.inode); + iput(whiteout_args.inode); + whiteout_args.inode = NULL; } } out_fail: - /* - * If we have pinned the log and an error happened, we unpin tasks - * trying to sync the log and force them to fallback to a transaction - * commit if the log currently contains any of the inodes involved in - * this rename operation (to ensure we do not persist a log with an - * inconsistent state for any of these inodes or leading to any - * inconsistencies when replayed). If the transaction was aborted, the - * abortion reason is propagated to userspace when attempting to commit - * the transaction. If the log does not contain any of these inodes, we - * allow the tasks to sync it. - */ - if (ret && log_pinned) { - 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(BTRFS_I(new_inode), fs_info->generation))) - btrfs_set_log_full_commit(trans); - - btrfs_end_log_trans(root); - log_pinned = false; - } - if (!ret && sync_log) { - ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, &ctx); - if (ret) - commit_transaction = true; - } else if (sync_log) { - mutex_lock(&root->log_mutex); - list_del(&ctx.list); - mutex_unlock(&root->log_mutex); - } - if (commit_transaction) { - ret = btrfs_commit_transaction(trans); - } else { - int ret2; - - ret2 = btrfs_end_transaction(trans); - ret = ret ? ret : ret2; - } + ret2 = btrfs_end_transaction(trans); + ret = ret ? ret : ret2; out_notrans: if (old_ino == BTRFS_FIRST_FREE_OBJECTID) up_read(&fs_info->subvol_sem); - + if (flags & RENAME_WHITEOUT) + btrfs_new_inode_args_destroy(&whiteout_args); +out_whiteout_inode: + if (flags & RENAME_WHITEOUT) + iput(whiteout_args.inode); return ret; } -static int btrfs_rename2(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int btrfs_rename2(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { + int ret; + if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT)) return -EINVAL; if (flags & RENAME_EXCHANGE) - return btrfs_rename_exchange(old_dir, old_dentry, new_dir, - new_dentry); + ret = btrfs_rename_exchange(old_dir, old_dentry, new_dir, + new_dentry); + else + ret = btrfs_rename(mnt_userns, old_dir, old_dentry, new_dir, + new_dentry, flags); + + btrfs_btree_balance_dirty(BTRFS_I(new_dir)->root->fs_info); - return btrfs_rename(old_dir, old_dentry, new_dir, new_dentry, flags); + return ret; } struct btrfs_delalloc_work { @@ -9574,7 +9544,9 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode * some fairly slow code that needs optimization. This walks the list * of all the inodes with pending delalloc and forces them to disk. */ -static int start_delalloc_inodes(struct btrfs_root *root, int nr, bool snapshot) +static int start_delalloc_inodes(struct btrfs_root *root, + struct writeback_control *wbc, bool snapshot, + bool in_reclaim_context) { struct btrfs_inode *binode; struct inode *inode; @@ -9582,6 +9554,7 @@ static int start_delalloc_inodes(struct btrfs_root *root, int nr, bool snapshot) struct list_head works; struct list_head splice; int ret = 0; + bool full_flush = wbc->nr_to_write == LONG_MAX; INIT_LIST_HEAD(&works); INIT_LIST_HEAD(&splice); @@ -9595,6 +9568,11 @@ static int start_delalloc_inodes(struct btrfs_root *root, int nr, bool snapshot) list_move_tail(&binode->delalloc_inodes, &root->delalloc_inodes); + + if (in_reclaim_context && + test_bit(BTRFS_INODE_NO_DELALLOC_FLUSH, &binode->runtime_flags)) + continue; + inode = igrab(&binode->vfs_inode); if (!inode) { cond_resched_lock(&root->delalloc_lock); @@ -9605,18 +9583,22 @@ static int start_delalloc_inodes(struct btrfs_root *root, int nr, bool snapshot) if (snapshot) set_bit(BTRFS_INODE_SNAPSHOT_FLUSH, &binode->runtime_flags); - work = btrfs_alloc_delalloc_work(inode); - if (!work) { - iput(inode); - ret = -ENOMEM; - goto out; + if (full_flush) { + work = btrfs_alloc_delalloc_work(inode); + if (!work) { + iput(inode); + ret = -ENOMEM; + goto out; + } + list_add_tail(&work->list, &works); + btrfs_queue_work(root->fs_info->flush_workers, + &work->work); + } else { + ret = filemap_fdatawrite_wbc(inode->i_mapping, wbc); + btrfs_add_delayed_iput(inode); + if (ret || wbc->nr_to_write <= 0) + goto out; } - list_add_tail(&work->list, &works); - btrfs_queue_work(root->fs_info->flush_workers, - &work->work); - ret++; - if (nr != -1 && ret >= nr) - goto out; cond_resched(); spin_lock(&root->delalloc_lock); } @@ -9638,27 +9620,36 @@ out: return ret; } -int btrfs_start_delalloc_snapshot(struct btrfs_root *root) +int btrfs_start_delalloc_snapshot(struct btrfs_root *root, bool in_reclaim_context) { + struct writeback_control wbc = { + .nr_to_write = LONG_MAX, + .sync_mode = WB_SYNC_NONE, + .range_start = 0, + .range_end = LLONG_MAX, + }; struct btrfs_fs_info *fs_info = root->fs_info; - int ret; - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + if (BTRFS_FS_ERROR(fs_info)) return -EROFS; - ret = start_delalloc_inodes(root, -1, true); - if (ret > 0) - ret = 0; - return ret; + return start_delalloc_inodes(root, &wbc, true, in_reclaim_context); } -int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr) +int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, long nr, + bool in_reclaim_context) { + struct writeback_control wbc = { + .nr_to_write = nr, + .sync_mode = WB_SYNC_NONE, + .range_start = 0, + .range_end = LLONG_MAX, + }; struct btrfs_root *root; struct list_head splice; int ret; - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + if (BTRFS_FS_ERROR(fs_info)) return -EROFS; INIT_LIST_HEAD(&splice); @@ -9666,24 +9657,26 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr) mutex_lock(&fs_info->delalloc_root_mutex); spin_lock(&fs_info->delalloc_root_lock); list_splice_init(&fs_info->delalloc_roots, &splice); - while (!list_empty(&splice) && nr) { + while (!list_empty(&splice)) { + /* + * Reset nr_to_write here so we know that we're doing a full + * flush. + */ + if (nr == LONG_MAX) + wbc.nr_to_write = LONG_MAX; + root = list_first_entry(&splice, struct btrfs_root, delalloc_root); - root = btrfs_grab_fs_root(root); + root = btrfs_grab_root(root); BUG_ON(!root); list_move_tail(&root->delalloc_root, &fs_info->delalloc_roots); spin_unlock(&fs_info->delalloc_root_lock); - ret = start_delalloc_inodes(root, nr, false); - btrfs_put_fs_root(root); - if (ret < 0) + ret = start_delalloc_inodes(root, &wbc, false, in_reclaim_context); + btrfs_put_root(root); + if (ret < 0 || wbc.nr_to_write <= 0) goto out; - - if (nr != -1) { - nr -= ret; - WARN_ON(nr < 0); - } spin_lock(&fs_info->delalloc_root_lock); } spin_unlock(&fs_info->delalloc_root_lock); @@ -9699,18 +9692,21 @@ out: return ret; } -static int btrfs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int btrfs_symlink(struct user_namespace *mnt_userns, 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; struct btrfs_key key; - struct inode *inode = NULL; + struct inode *inode; + struct btrfs_new_inode_args new_inode_args = { + .dir = dir, + .dentry = dentry, + }; + unsigned int trans_num_items; int err; - u64 objectid; - u64 index = 0; int name_len; int datasize; unsigned long ptr; @@ -9721,49 +9717,40 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info)) return -ENAMETOOLONG; - /* - * 2 items for inode item and ref - * 2 items for dir items - * 1 item for updating parent inode item - * 1 item for the inline extent item - * 1 item for xattr if selinux is on - */ - trans = btrfs_start_transaction(root, 7); - if (IS_ERR(trans)) - return PTR_ERR(trans); + inode = new_inode(dir->i_sb); + if (!inode) + return -ENOMEM; + inode_init_owner(mnt_userns, inode, dir, S_IFLNK | S_IRWXUGO); + inode->i_op = &btrfs_symlink_inode_operations; + inode_nohighmem(inode); + inode->i_mapping->a_ops = &btrfs_aops; + btrfs_i_size_write(BTRFS_I(inode), name_len); + inode_set_bytes(inode, name_len); - err = btrfs_find_free_ino(root, &objectid); + new_inode_args.inode = inode; + err = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items); if (err) - goto out_unlock; + goto out_inode; + /* 1 additional item for the inline extent */ + trans_num_items++; - inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), - objectid, S_IFLNK|S_IRWXUGO, &index); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - inode = NULL; - goto out_unlock; + trans = btrfs_start_transaction(root, trans_num_items); + if (IS_ERR(trans)) { + err = PTR_ERR(trans); + goto out_new_inode_args; } - /* - * If the active LSM wants to access the inode during - * d_instantiate it needs these. Smack checks to see - * if the filesystem supports xattrs by looking at the - * ops vector. - */ - inode->i_fop = &btrfs_file_operations; - inode->i_op = &btrfs_file_inode_operations; - inode->i_mapping->a_ops = &btrfs_aops; - BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - - err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); + err = btrfs_create_new_inode(trans, &new_inode_args); if (err) - goto out_unlock; + goto out; path = btrfs_alloc_path(); if (!path) { err = -ENOMEM; - goto out_unlock; + btrfs_abort_transaction(trans, err); + discard_new_inode(inode); + inode = NULL; + goto out; } key.objectid = btrfs_ino(BTRFS_I(inode)); key.offset = 0; @@ -9772,8 +9759,11 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, err = btrfs_insert_empty_item(trans, root, path, &key, datasize); if (err) { + btrfs_abort_transaction(trans, err); btrfs_free_path(path); - goto out_unlock; + discard_new_inode(inode); + inode = NULL; + goto out; } leaf = path->nodes[0]; ei = btrfs_item_ptr(leaf, path->slots[0], @@ -9791,41 +9781,102 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, btrfs_mark_buffer_dirty(leaf); btrfs_free_path(path); - inode->i_op = &btrfs_symlink_inode_operations; - inode_nohighmem(inode); - inode_set_bytes(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 - * last step to avoid extra cleanup of these indexes if an error happens - * elsewhere above. - */ - if (!err) - err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, - BTRFS_I(inode), 0, index); - if (err) - goto out_unlock; - d_instantiate_new(dentry, inode); - -out_unlock: + err = 0; +out: btrfs_end_transaction(trans); - if (err && inode) { - inode_dec_link_count(inode); - discard_new_inode(inode); - } btrfs_btree_balance_dirty(fs_info); +out_new_inode_args: + btrfs_new_inode_args_destroy(&new_inode_args); +out_inode: + if (err) + iput(inode); return err; } +static struct btrfs_trans_handle *insert_prealloc_file_extent( + struct btrfs_trans_handle *trans_in, + struct btrfs_inode *inode, + struct btrfs_key *ins, + u64 file_offset) +{ + struct btrfs_file_extent_item stack_fi; + struct btrfs_replace_extent_info extent_info; + struct btrfs_trans_handle *trans = trans_in; + struct btrfs_path *path; + u64 start = ins->objectid; + u64 len = ins->offset; + int qgroup_released; + int ret; + + memset(&stack_fi, 0, sizeof(stack_fi)); + + btrfs_set_stack_file_extent_type(&stack_fi, BTRFS_FILE_EXTENT_PREALLOC); + btrfs_set_stack_file_extent_disk_bytenr(&stack_fi, start); + btrfs_set_stack_file_extent_disk_num_bytes(&stack_fi, len); + btrfs_set_stack_file_extent_num_bytes(&stack_fi, len); + btrfs_set_stack_file_extent_ram_bytes(&stack_fi, len); + btrfs_set_stack_file_extent_compression(&stack_fi, BTRFS_COMPRESS_NONE); + /* Encryption and other encoding is reserved and all 0 */ + + qgroup_released = btrfs_qgroup_release_data(inode, file_offset, len); + if (qgroup_released < 0) + return ERR_PTR(qgroup_released); + + if (trans) { + ret = insert_reserved_file_extent(trans, inode, + file_offset, &stack_fi, + true, qgroup_released); + if (ret) + goto free_qgroup; + return trans; + } + + extent_info.disk_offset = start; + extent_info.disk_len = len; + extent_info.data_offset = 0; + extent_info.data_len = len; + extent_info.file_offset = file_offset; + extent_info.extent_buf = (char *)&stack_fi; + extent_info.is_new_extent = true; + extent_info.update_times = true; + extent_info.qgroup_reserved = qgroup_released; + extent_info.insertions = 0; + + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto free_qgroup; + } + + ret = btrfs_replace_file_extents(inode, path, file_offset, + file_offset + len - 1, &extent_info, + &trans); + btrfs_free_path(path); + if (ret) + goto free_qgroup; + return trans; + +free_qgroup: + /* + * We have released qgroup data range at the beginning of the function, + * and normally qgroup_released bytes will be freed when committing + * transaction. + * But if we error out early, we have to free what we have released + * or we leak qgroup data reservation. + */ + btrfs_qgroup_free_refroot(inode->root->fs_info, + inode->root->root_key.objectid, qgroup_released, + BTRFS_QGROUP_RSV_DATA); + return ERR_PTR(ret); +} + static int __btrfs_prealloc_file_range(struct inode *inode, int mode, u64 start, u64 num_bytes, u64 min_size, 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; struct btrfs_key ins; @@ -9841,14 +9892,6 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, if (trans) own_trans = false; while (num_bytes > 0) { - if (own_trans) { - trans = btrfs_start_transaction(root, 3); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - break; - } - } - cur_bytes = min_t(u64, num_bytes, SZ_256M); cur_bytes = max(cur_bytes, min_size); /* @@ -9860,11 +9903,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, cur_bytes = min(cur_bytes, last_alloc); ret = btrfs_reserve_extent(root, cur_bytes, cur_bytes, min_size, 0, *alloc_hint, &ins, 1, 0); - if (ret) { - if (own_trans) - btrfs_end_transaction(trans); + if (ret) break; - } /* * We've reserved this space, and thus converted it from @@ -9874,30 +9914,29 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, * clear_offset by our extent size. */ clear_offset += ins.offset; - btrfs_dec_block_group_reservations(fs_info, ins.objectid); last_alloc = ins.offset; - ret = insert_reserved_file_extent(trans, inode, - cur_offset, ins.objectid, - ins.offset, ins.offset, - ins.offset, 0, 0, 0, - BTRFS_FILE_EXTENT_PREALLOC); - if (ret) { + trans = insert_prealloc_file_extent(trans, BTRFS_I(inode), + &ins, cur_offset); + /* + * Now that we inserted the prealloc extent we can finally + * decrement the number of reservations in the block group. + * If we did it before, we could race with relocation and have + * relocation miss the reserved extent, making it fail later. + */ + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 0); - btrfs_abort_transaction(trans, ret); - if (own_trans) - btrfs_end_transaction(trans); break; } - btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, - cur_offset + ins.offset -1, 0); - em = alloc_extent_map(); if (!em) { - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &BTRFS_I(inode)->runtime_flags); + btrfs_drop_extent_map_range(BTRFS_I(inode), cur_offset, + cur_offset + ins.offset - 1, false); + btrfs_set_inode_full_sync(BTRFS_I(inode)); goto next; } @@ -9911,16 +9950,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, set_bit(EXTENT_FLAG_PREALLOC, &em->flags); em->generation = trans->transid; - while (1) { - write_lock(&em_tree->lock); - ret = add_extent_mapping(em_tree, em, 1); - write_unlock(&em_tree->lock); - if (ret != -EEXIST) - break; - btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, - cur_offset + ins.offset - 1, - 0); - } + ret = btrfs_replace_extent_map_range(BTRFS_I(inode), em, true); free_extent_map(em); next: num_bytes -= ins.offset; @@ -9938,10 +9968,10 @@ next: else i_size = cur_offset; i_size_write(inode, i_size); - btrfs_ordered_update_i_size(inode, i_size, NULL); + btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0); } - ret = btrfs_update_inode(trans, root, inode); + ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); if (ret) { btrfs_abort_transaction(trans, ret); @@ -9950,11 +9980,13 @@ next: break; } - if (own_trans) + if (own_trans) { btrfs_end_transaction(trans); + trans = NULL; + } } if (clear_offset < end) - btrfs_free_reserved_data_space(inode, NULL, clear_offset, + btrfs_free_reserved_data_space(BTRFS_I(inode), NULL, clear_offset, end - clear_offset + 1); return ret; } @@ -9977,12 +10009,8 @@ int btrfs_prealloc_file_range_trans(struct inode *inode, min_size, actual_len, alloc_hint, trans); } -static int btrfs_set_page_dirty(struct page *page) -{ - return __set_page_dirty_nobuffers(page); -} - -static int btrfs_permission(struct inode *inode, int mask) +static int btrfs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct btrfs_root *root = BTRFS_I(inode)->root; umode_t mode = inode->i_mode; @@ -9994,90 +10022,813 @@ static int btrfs_permission(struct inode *inode, int mask) if (BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) return -EACCES; } - return generic_permission(inode, mask); + return generic_permission(mnt_userns, inode, mask); } -static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int btrfs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct file *file, 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; - u64 objectid; - u64 index; - int ret = 0; - - /* - * 5 units required for adding orphan entry - */ - trans = btrfs_start_transaction(root, 5); - if (IS_ERR(trans)) - return PTR_ERR(trans); - - ret = btrfs_find_free_ino(root, &objectid); - if (ret) - goto out; - - inode = btrfs_new_inode(trans, root, dir, NULL, 0, - btrfs_ino(BTRFS_I(dir)), objectid, mode, &index); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - inode = NULL; - goto out; - } + struct inode *inode; + struct btrfs_new_inode_args new_inode_args = { + .dir = dir, + .dentry = file->f_path.dentry, + .orphan = true, + }; + unsigned int trans_num_items; + int ret; + inode = new_inode(dir->i_sb); + if (!inode) + return -ENOMEM; + inode_init_owner(mnt_userns, inode, dir, mode); inode->i_fop = &btrfs_file_operations; inode->i_op = &btrfs_file_inode_operations; - inode->i_mapping->a_ops = &btrfs_aops; - BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - ret = btrfs_init_inode_security(trans, inode, dir, NULL); + new_inode_args.inode = inode; + ret = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items); if (ret) - goto out; + goto out_inode; - ret = btrfs_update_inode(trans, root, inode); - if (ret) - goto out; - ret = btrfs_orphan_add(trans, BTRFS_I(inode)); - if (ret) - goto out; + trans = btrfs_start_transaction(root, trans_num_items); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out_new_inode_args; + } + + ret = btrfs_create_new_inode(trans, &new_inode_args); /* - * We set number of links to 0 in btrfs_new_inode(), and here we set - * it to 1 because d_tmpfile() will issue a warning if the count is 0, - * through: + * We set number of links to 0 in btrfs_create_new_inode(), and here we + * set it to 1 because d_tmpfile() will issue a warning if the count is + * 0, through: * * d_tmpfile() -> inode_dec_link_count() -> drop_nlink() */ set_nlink(inode, 1); - d_tmpfile(dentry, inode); - unlock_new_inode(inode); - mark_inode_dirty(inode); -out: + + if (!ret) { + d_tmpfile(file, inode); + unlock_new_inode(inode); + mark_inode_dirty(inode); + } + btrfs_end_transaction(trans); - if (ret && inode) - discard_new_inode(inode); btrfs_btree_balance_dirty(fs_info); - return ret; +out_new_inode_args: + btrfs_new_inode_args_destroy(&new_inode_args); +out_inode: + if (ret) + iput(inode); + return finish_open_simple(file, ret); } -void btrfs_set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end) +void btrfs_set_range_writeback(struct btrfs_inode *inode, u64 start, u64 end) { - struct inode *inode = tree->private_data; + struct btrfs_fs_info *fs_info = inode->root->fs_info; unsigned long index = start >> PAGE_SHIFT; unsigned long end_index = end >> PAGE_SHIFT; struct page *page; + u32 len; + ASSERT(end + 1 - start <= U32_MAX); + len = end + 1 - start; while (index <= end_index) { - page = find_get_page(inode->i_mapping, index); + page = find_get_page(inode->vfs_inode.i_mapping, index); ASSERT(page); /* Pages should be in the extent_io_tree */ - set_page_writeback(page); + + btrfs_page_set_writeback(fs_info, page, start, len); put_page(page); index++; } } +int btrfs_encoded_io_compression_from_extent(struct btrfs_fs_info *fs_info, + int compress_type) +{ + switch (compress_type) { + case BTRFS_COMPRESS_NONE: + return BTRFS_ENCODED_IO_COMPRESSION_NONE; + case BTRFS_COMPRESS_ZLIB: + return BTRFS_ENCODED_IO_COMPRESSION_ZLIB; + case BTRFS_COMPRESS_LZO: + /* + * The LZO format depends on the sector size. 64K is the maximum + * sector size that we support. + */ + if (fs_info->sectorsize < SZ_4K || fs_info->sectorsize > SZ_64K) + return -EINVAL; + return BTRFS_ENCODED_IO_COMPRESSION_LZO_4K + + (fs_info->sectorsize_bits - 12); + case BTRFS_COMPRESS_ZSTD: + return BTRFS_ENCODED_IO_COMPRESSION_ZSTD; + default: + return -EUCLEAN; + } +} + +static ssize_t btrfs_encoded_read_inline( + struct kiocb *iocb, + struct iov_iter *iter, u64 start, + u64 lockend, + struct extent_state **cached_state, + u64 extent_start, size_t count, + struct btrfs_ioctl_encoded_io_args *encoded, + bool *unlocked) +{ + struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_io_tree *io_tree = &inode->io_tree; + struct btrfs_path *path; + struct extent_buffer *leaf; + struct btrfs_file_extent_item *item; + u64 ram_bytes; + unsigned long ptr; + void *tmp; + ssize_t ret; + + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto out; + } + ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), + extent_start, 0); + if (ret) { + if (ret > 0) { + /* The extent item disappeared? */ + ret = -EIO; + } + goto out; + } + leaf = path->nodes[0]; + item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + + ram_bytes = btrfs_file_extent_ram_bytes(leaf, item); + ptr = btrfs_file_extent_inline_start(item); + + encoded->len = min_t(u64, extent_start + ram_bytes, + inode->vfs_inode.i_size) - iocb->ki_pos; + ret = btrfs_encoded_io_compression_from_extent(fs_info, + btrfs_file_extent_compression(leaf, item)); + if (ret < 0) + goto out; + encoded->compression = ret; + if (encoded->compression) { + size_t inline_size; + + inline_size = btrfs_file_extent_inline_item_len(leaf, + path->slots[0]); + if (inline_size > count) { + ret = -ENOBUFS; + goto out; + } + count = inline_size; + encoded->unencoded_len = ram_bytes; + encoded->unencoded_offset = iocb->ki_pos - extent_start; + } else { + count = min_t(u64, count, encoded->len); + encoded->len = count; + encoded->unencoded_len = count; + ptr += iocb->ki_pos - extent_start; + } + + tmp = kmalloc(count, GFP_NOFS); + if (!tmp) { + ret = -ENOMEM; + goto out; + } + read_extent_buffer(leaf, tmp, ptr, count); + btrfs_release_path(path); + unlock_extent(io_tree, start, lockend, cached_state); + btrfs_inode_unlock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + *unlocked = true; + + ret = copy_to_iter(tmp, count, iter); + if (ret != count) + ret = -EFAULT; + kfree(tmp); +out: + btrfs_free_path(path); + return ret; +} + +struct btrfs_encoded_read_private { + struct btrfs_inode *inode; + u64 file_offset; + wait_queue_head_t wait; + atomic_t pending; + blk_status_t status; + bool skip_csum; +}; + +static blk_status_t submit_encoded_read_bio(struct btrfs_inode *inode, + struct bio *bio, int mirror_num) +{ + struct btrfs_encoded_read_private *priv = btrfs_bio(bio)->private; + struct btrfs_fs_info *fs_info = inode->root->fs_info; + blk_status_t ret; + + if (!priv->skip_csum) { + ret = btrfs_lookup_bio_sums(&inode->vfs_inode, bio, NULL); + if (ret) + return ret; + } + + atomic_inc(&priv->pending); + btrfs_submit_bio(fs_info, bio, mirror_num); + return BLK_STS_OK; +} + +static blk_status_t btrfs_encoded_read_verify_csum(struct btrfs_bio *bbio) +{ + const bool uptodate = (bbio->bio.bi_status == BLK_STS_OK); + struct btrfs_encoded_read_private *priv = bbio->private; + struct btrfs_inode *inode = priv->inode; + struct btrfs_fs_info *fs_info = inode->root->fs_info; + u32 sectorsize = fs_info->sectorsize; + struct bio_vec *bvec; + struct bvec_iter_all iter_all; + u32 bio_offset = 0; + + if (priv->skip_csum || !uptodate) + return bbio->bio.bi_status; + + bio_for_each_segment_all(bvec, &bbio->bio, iter_all) { + unsigned int i, nr_sectors, pgoff; + + nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len); + pgoff = bvec->bv_offset; + for (i = 0; i < nr_sectors; i++) { + ASSERT(pgoff < PAGE_SIZE); + if (btrfs_check_data_csum(&inode->vfs_inode, bbio, bio_offset, + bvec->bv_page, pgoff)) + return BLK_STS_IOERR; + bio_offset += sectorsize; + pgoff += sectorsize; + } + } + return BLK_STS_OK; +} + +static void btrfs_encoded_read_endio(struct btrfs_bio *bbio) +{ + struct btrfs_encoded_read_private *priv = bbio->private; + blk_status_t status; + + status = btrfs_encoded_read_verify_csum(bbio); + if (status) { + /* + * The memory barrier implied by the atomic_dec_return() here + * pairs with the memory barrier implied by the + * atomic_dec_return() or io_wait_event() in + * btrfs_encoded_read_regular_fill_pages() to ensure that this + * write is observed before the load of status in + * btrfs_encoded_read_regular_fill_pages(). + */ + WRITE_ONCE(priv->status, status); + } + if (!atomic_dec_return(&priv->pending)) + wake_up(&priv->wait); + btrfs_bio_free_csum(bbio); + bio_put(&bbio->bio); +} + +int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode, + u64 file_offset, u64 disk_bytenr, + u64 disk_io_size, struct page **pages) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct btrfs_encoded_read_private priv = { + .inode = inode, + .file_offset = file_offset, + .pending = ATOMIC_INIT(1), + .skip_csum = (inode->flags & BTRFS_INODE_NODATASUM), + }; + unsigned long i = 0; + u64 cur = 0; + int ret; + + init_waitqueue_head(&priv.wait); + /* + * Submit bios for the extent, splitting due to bio or stripe limits as + * necessary. + */ + while (cur < disk_io_size) { + struct extent_map *em; + struct btrfs_io_geometry geom; + struct bio *bio = NULL; + u64 remaining; + + em = btrfs_get_chunk_map(fs_info, disk_bytenr + cur, + disk_io_size - cur); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + } else { + ret = btrfs_get_io_geometry(fs_info, em, BTRFS_MAP_READ, + disk_bytenr + cur, &geom); + free_extent_map(em); + } + if (ret) { + WRITE_ONCE(priv.status, errno_to_blk_status(ret)); + break; + } + remaining = min(geom.len, disk_io_size - cur); + while (bio || remaining) { + size_t bytes = min_t(u64, remaining, PAGE_SIZE); + + if (!bio) { + bio = btrfs_bio_alloc(BIO_MAX_VECS, REQ_OP_READ, + btrfs_encoded_read_endio, + &priv); + bio->bi_iter.bi_sector = + (disk_bytenr + cur) >> SECTOR_SHIFT; + } + + if (!bytes || + bio_add_page(bio, pages[i], bytes, 0) < bytes) { + blk_status_t status; + + status = submit_encoded_read_bio(inode, bio, 0); + if (status) { + WRITE_ONCE(priv.status, status); + bio_put(bio); + goto out; + } + bio = NULL; + continue; + } + + i++; + cur += bytes; + remaining -= bytes; + } + } + +out: + if (atomic_dec_return(&priv.pending)) + io_wait_event(priv.wait, !atomic_read(&priv.pending)); + /* See btrfs_encoded_read_endio() for ordering. */ + return blk_status_to_errno(READ_ONCE(priv.status)); +} + +static ssize_t btrfs_encoded_read_regular(struct kiocb *iocb, + struct iov_iter *iter, + u64 start, u64 lockend, + struct extent_state **cached_state, + u64 disk_bytenr, u64 disk_io_size, + size_t count, bool compressed, + bool *unlocked) +{ + struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); + struct extent_io_tree *io_tree = &inode->io_tree; + struct page **pages; + unsigned long nr_pages, i; + u64 cur; + size_t page_offset; + ssize_t ret; + + nr_pages = DIV_ROUND_UP(disk_io_size, PAGE_SIZE); + pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS); + if (!pages) + return -ENOMEM; + ret = btrfs_alloc_page_array(nr_pages, pages); + if (ret) { + ret = -ENOMEM; + goto out; + } + + ret = btrfs_encoded_read_regular_fill_pages(inode, start, disk_bytenr, + disk_io_size, pages); + if (ret) + goto out; + + unlock_extent(io_tree, start, lockend, cached_state); + btrfs_inode_unlock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + *unlocked = true; + + if (compressed) { + i = 0; + page_offset = 0; + } else { + i = (iocb->ki_pos - start) >> PAGE_SHIFT; + page_offset = (iocb->ki_pos - start) & (PAGE_SIZE - 1); + } + cur = 0; + while (cur < count) { + size_t bytes = min_t(size_t, count - cur, + PAGE_SIZE - page_offset); + + if (copy_page_to_iter(pages[i], page_offset, bytes, + iter) != bytes) { + ret = -EFAULT; + goto out; + } + i++; + cur += bytes; + page_offset = 0; + } + ret = count; +out: + for (i = 0; i < nr_pages; i++) { + if (pages[i]) + __free_page(pages[i]); + } + kfree(pages); + return ret; +} + +ssize_t btrfs_encoded_read(struct kiocb *iocb, struct iov_iter *iter, + struct btrfs_ioctl_encoded_io_args *encoded) +{ + struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_io_tree *io_tree = &inode->io_tree; + ssize_t ret; + size_t count = iov_iter_count(iter); + u64 start, lockend, disk_bytenr, disk_io_size; + struct extent_state *cached_state = NULL; + struct extent_map *em; + bool unlocked = false; + + file_accessed(iocb->ki_filp); + + btrfs_inode_lock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + + if (iocb->ki_pos >= inode->vfs_inode.i_size) { + btrfs_inode_unlock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + return 0; + } + start = ALIGN_DOWN(iocb->ki_pos, fs_info->sectorsize); + /* + * We don't know how long the extent containing iocb->ki_pos is, but if + * it's compressed we know that it won't be longer than this. + */ + lockend = start + BTRFS_MAX_UNCOMPRESSED - 1; + + for (;;) { + struct btrfs_ordered_extent *ordered; + + ret = btrfs_wait_ordered_range(&inode->vfs_inode, start, + lockend - start + 1); + if (ret) + goto out_unlock_inode; + lock_extent(io_tree, start, lockend, &cached_state); + ordered = btrfs_lookup_ordered_range(inode, start, + lockend - start + 1); + if (!ordered) + break; + btrfs_put_ordered_extent(ordered); + unlock_extent(io_tree, start, lockend, &cached_state); + cond_resched(); + } + + em = btrfs_get_extent(inode, NULL, 0, start, lockend - start + 1); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + goto out_unlock_extent; + } + + if (em->block_start == EXTENT_MAP_INLINE) { + u64 extent_start = em->start; + + /* + * For inline extents we get everything we need out of the + * extent item. + */ + free_extent_map(em); + em = NULL; + ret = btrfs_encoded_read_inline(iocb, iter, start, lockend, + &cached_state, extent_start, + count, encoded, &unlocked); + goto out; + } + + /* + * We only want to return up to EOF even if the extent extends beyond + * that. + */ + encoded->len = min_t(u64, extent_map_end(em), + inode->vfs_inode.i_size) - iocb->ki_pos; + if (em->block_start == EXTENT_MAP_HOLE || + test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { + disk_bytenr = EXTENT_MAP_HOLE; + count = min_t(u64, count, encoded->len); + encoded->len = count; + encoded->unencoded_len = count; + } else if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) { + disk_bytenr = em->block_start; + /* + * Bail if the buffer isn't large enough to return the whole + * compressed extent. + */ + if (em->block_len > count) { + ret = -ENOBUFS; + goto out_em; + } + disk_io_size = em->block_len; + count = em->block_len; + encoded->unencoded_len = em->ram_bytes; + encoded->unencoded_offset = iocb->ki_pos - em->orig_start; + ret = btrfs_encoded_io_compression_from_extent(fs_info, + em->compress_type); + if (ret < 0) + goto out_em; + encoded->compression = ret; + } else { + disk_bytenr = em->block_start + (start - em->start); + if (encoded->len > count) + encoded->len = count; + /* + * Don't read beyond what we locked. This also limits the page + * allocations that we'll do. + */ + disk_io_size = min(lockend + 1, iocb->ki_pos + encoded->len) - start; + count = start + disk_io_size - iocb->ki_pos; + encoded->len = count; + encoded->unencoded_len = count; + disk_io_size = ALIGN(disk_io_size, fs_info->sectorsize); + } + free_extent_map(em); + em = NULL; + + if (disk_bytenr == EXTENT_MAP_HOLE) { + unlock_extent(io_tree, start, lockend, &cached_state); + btrfs_inode_unlock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + unlocked = true; + ret = iov_iter_zero(count, iter); + if (ret != count) + ret = -EFAULT; + } else { + ret = btrfs_encoded_read_regular(iocb, iter, start, lockend, + &cached_state, disk_bytenr, + disk_io_size, count, + encoded->compression, + &unlocked); + } + +out: + if (ret >= 0) + iocb->ki_pos += encoded->len; +out_em: + free_extent_map(em); +out_unlock_extent: + if (!unlocked) + unlock_extent(io_tree, start, lockend, &cached_state); +out_unlock_inode: + if (!unlocked) + btrfs_inode_unlock(&inode->vfs_inode, BTRFS_ILOCK_SHARED); + return ret; +} + +ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, + const struct btrfs_ioctl_encoded_io_args *encoded) +{ + struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_io_tree *io_tree = &inode->io_tree; + struct extent_changeset *data_reserved = NULL; + struct extent_state *cached_state = NULL; + int compression; + size_t orig_count; + u64 start, end; + u64 num_bytes, ram_bytes, disk_num_bytes; + unsigned long nr_pages, i; + struct page **pages; + struct btrfs_key ins; + bool extent_reserved = false; + struct extent_map *em; + ssize_t ret; + + switch (encoded->compression) { + case BTRFS_ENCODED_IO_COMPRESSION_ZLIB: + compression = BTRFS_COMPRESS_ZLIB; + break; + case BTRFS_ENCODED_IO_COMPRESSION_ZSTD: + compression = BTRFS_COMPRESS_ZSTD; + break; + case BTRFS_ENCODED_IO_COMPRESSION_LZO_4K: + case BTRFS_ENCODED_IO_COMPRESSION_LZO_8K: + case BTRFS_ENCODED_IO_COMPRESSION_LZO_16K: + case BTRFS_ENCODED_IO_COMPRESSION_LZO_32K: + case BTRFS_ENCODED_IO_COMPRESSION_LZO_64K: + /* The sector size must match for LZO. */ + if (encoded->compression - + BTRFS_ENCODED_IO_COMPRESSION_LZO_4K + 12 != + fs_info->sectorsize_bits) + return -EINVAL; + compression = BTRFS_COMPRESS_LZO; + break; + default: + return -EINVAL; + } + if (encoded->encryption != BTRFS_ENCODED_IO_ENCRYPTION_NONE) + return -EINVAL; + + orig_count = iov_iter_count(from); + + /* The extent size must be sane. */ + if (encoded->unencoded_len > BTRFS_MAX_UNCOMPRESSED || + orig_count > BTRFS_MAX_COMPRESSED || orig_count == 0) + return -EINVAL; + + /* + * The compressed data must be smaller than the decompressed data. + * + * It's of course possible for data to compress to larger or the same + * size, but the buffered I/O path falls back to no compression for such + * data, and we don't want to break any assumptions by creating these + * extents. + * + * Note that this is less strict than the current check we have that the + * compressed data must be at least one sector smaller than the + * decompressed data. We only want to enforce the weaker requirement + * from old kernels that it is at least one byte smaller. + */ + if (orig_count >= encoded->unencoded_len) + return -EINVAL; + + /* The extent must start on a sector boundary. */ + start = iocb->ki_pos; + if (!IS_ALIGNED(start, fs_info->sectorsize)) + return -EINVAL; + + /* + * The extent must end on a sector boundary. However, we allow a write + * which ends at or extends i_size to have an unaligned length; we round + * up the extent size and set i_size to the unaligned end. + */ + if (start + encoded->len < inode->vfs_inode.i_size && + !IS_ALIGNED(start + encoded->len, fs_info->sectorsize)) + return -EINVAL; + + /* Finally, the offset in the unencoded data must be sector-aligned. */ + if (!IS_ALIGNED(encoded->unencoded_offset, fs_info->sectorsize)) + return -EINVAL; + + num_bytes = ALIGN(encoded->len, fs_info->sectorsize); + ram_bytes = ALIGN(encoded->unencoded_len, fs_info->sectorsize); + end = start + num_bytes - 1; + + /* + * If the extent cannot be inline, the compressed data on disk must be + * sector-aligned. For convenience, we extend it with zeroes if it + * isn't. + */ + disk_num_bytes = ALIGN(orig_count, fs_info->sectorsize); + nr_pages = DIV_ROUND_UP(disk_num_bytes, PAGE_SIZE); + pages = kvcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL_ACCOUNT); + if (!pages) + return -ENOMEM; + for (i = 0; i < nr_pages; i++) { + size_t bytes = min_t(size_t, PAGE_SIZE, iov_iter_count(from)); + char *kaddr; + + pages[i] = alloc_page(GFP_KERNEL_ACCOUNT); + if (!pages[i]) { + ret = -ENOMEM; + goto out_pages; + } + kaddr = kmap_local_page(pages[i]); + if (copy_from_iter(kaddr, bytes, from) != bytes) { + kunmap_local(kaddr); + ret = -EFAULT; + goto out_pages; + } + if (bytes < PAGE_SIZE) + memset(kaddr + bytes, 0, PAGE_SIZE - bytes); + kunmap_local(kaddr); + } + + for (;;) { + struct btrfs_ordered_extent *ordered; + + ret = btrfs_wait_ordered_range(&inode->vfs_inode, start, num_bytes); + if (ret) + goto out_pages; + ret = invalidate_inode_pages2_range(inode->vfs_inode.i_mapping, + start >> PAGE_SHIFT, + end >> PAGE_SHIFT); + if (ret) + goto out_pages; + lock_extent(io_tree, start, end, &cached_state); + ordered = btrfs_lookup_ordered_range(inode, start, num_bytes); + if (!ordered && + !filemap_range_has_page(inode->vfs_inode.i_mapping, start, end)) + break; + if (ordered) + btrfs_put_ordered_extent(ordered); + unlock_extent(io_tree, start, end, &cached_state); + cond_resched(); + } + + /* + * We don't use the higher-level delalloc space functions because our + * num_bytes and disk_num_bytes are different. + */ + ret = btrfs_alloc_data_chunk_ondemand(inode, disk_num_bytes); + if (ret) + goto out_unlock; + ret = btrfs_qgroup_reserve_data(inode, &data_reserved, start, num_bytes); + if (ret) + goto out_free_data_space; + ret = btrfs_delalloc_reserve_metadata(inode, num_bytes, disk_num_bytes, + false); + if (ret) + goto out_qgroup_free_data; + + /* Try an inline extent first. */ + if (start == 0 && encoded->unencoded_len == encoded->len && + encoded->unencoded_offset == 0) { + ret = cow_file_range_inline(inode, encoded->len, orig_count, + compression, pages, true); + if (ret <= 0) { + if (ret == 0) + ret = orig_count; + goto out_delalloc_release; + } + } + + ret = btrfs_reserve_extent(root, disk_num_bytes, disk_num_bytes, + disk_num_bytes, 0, 0, &ins, 1, 1); + if (ret) + goto out_delalloc_release; + extent_reserved = true; + + em = create_io_em(inode, start, num_bytes, + start - encoded->unencoded_offset, ins.objectid, + ins.offset, ins.offset, ram_bytes, compression, + BTRFS_ORDERED_COMPRESSED); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + goto out_free_reserved; + } + free_extent_map(em); + + ret = btrfs_add_ordered_extent(inode, start, num_bytes, ram_bytes, + ins.objectid, ins.offset, + encoded->unencoded_offset, + (1 << BTRFS_ORDERED_ENCODED) | + (1 << BTRFS_ORDERED_COMPRESSED), + compression); + if (ret) { + btrfs_drop_extent_map_range(inode, start, end, false); + goto out_free_reserved; + } + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + + if (start + encoded->len > inode->vfs_inode.i_size) + i_size_write(&inode->vfs_inode, start + encoded->len); + + unlock_extent(io_tree, start, end, &cached_state); + + btrfs_delalloc_release_extents(inode, num_bytes); + + if (btrfs_submit_compressed_write(inode, start, num_bytes, ins.objectid, + ins.offset, pages, nr_pages, 0, NULL, + false)) { + btrfs_writepage_endio_finish_ordered(inode, pages[0], start, end, 0); + ret = -EIO; + goto out_pages; + } + ret = orig_count; + goto out; + +out_free_reserved: + btrfs_dec_block_group_reservations(fs_info, ins.objectid); + btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); +out_delalloc_release: + btrfs_delalloc_release_extents(inode, num_bytes); + btrfs_delalloc_release_metadata(inode, disk_num_bytes, ret < 0); +out_qgroup_free_data: + if (ret < 0) + btrfs_qgroup_free_data(inode, data_reserved, start, num_bytes); +out_free_data_space: + /* + * If btrfs_reserve_extent() succeeded, then we already decremented + * bytes_may_use. + */ + if (!extent_reserved) + btrfs_free_reserved_data_space_noquota(fs_info, disk_num_bytes); +out_unlock: + unlock_extent(io_tree, start, end, &cached_state); +out_pages: + for (i = 0; i < nr_pages; i++) { + if (pages[i]) + __free_page(pages[i]); + } + kvfree(pages); +out: + if (ret >= 0) + iocb->ki_pos += encoded->len; + return ret; +} + #ifdef CONFIG_SWAP /* * Add an entry indicating a block group or device which is pinned by a @@ -10098,6 +10849,7 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr, sp->ptr = ptr; sp->inode = inode; sp->is_block_group = is_block_group; + sp->bg_extent_count = 1; spin_lock(&fs_info->swapfile_pins_lock); p = &fs_info->swapfile_pins.rb_node; @@ -10111,6 +10863,8 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr, (sp->ptr == entry->ptr && sp->inode > entry->inode)) { p = &(*p)->rb_right; } else { + if (is_block_group) + entry->bg_extent_count++; spin_unlock(&fs_info->swapfile_pins_lock); kfree(sp); return 1; @@ -10136,8 +10890,11 @@ static void btrfs_free_swapfile_pins(struct inode *inode) sp = rb_entry(node, struct btrfs_swapfile_pin, node); if (sp->inode == inode) { rb_erase(&sp->node, &fs_info->swapfile_pins); - if (sp->is_block_group) + if (sp->is_block_group) { + btrfs_dec_block_group_swap_extents(sp->ptr, + sp->bg_extent_count); btrfs_put_block_group(sp->ptr); + } kfree(sp); } node = next; @@ -10159,9 +10916,19 @@ static int btrfs_add_swap_extent(struct swap_info_struct *sis, struct btrfs_swap_info *bsi) { unsigned long nr_pages; + unsigned long max_pages; u64 first_ppage, first_ppage_reported, next_ppage; int ret; + /* + * Our swapfile may have had its size extended after the swap header was + * written. In that case activating the swapfile should not go beyond + * the max size set in the swap header. + */ + if (bsi->nr_pages >= sis->max) + return 0; + + max_pages = sis->max - bsi->nr_pages; first_ppage = ALIGN(bsi->block_start, PAGE_SIZE) >> PAGE_SHIFT; next_ppage = ALIGN_DOWN(bsi->block_start + bsi->block_len, PAGE_SIZE) >> PAGE_SHIFT; @@ -10169,6 +10936,7 @@ static int btrfs_add_swap_extent(struct swap_info_struct *sis, if (first_ppage >= next_ppage) return 0; nr_pages = next_ppage - first_ppage; + nr_pages = min(nr_pages, max_pages); first_ppage_reported = first_ppage; if (bsi->start == 0) @@ -10198,7 +10966,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, sector_t *span) { struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct extent_state *cached_state = NULL; struct extent_map *em = NULL; @@ -10237,29 +11006,58 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, /* * Balance or device remove/replace/resize can move stuff around from - * under us. The EXCL_OP flag makes sure they aren't running/won't run - * concurrently while we are mapping the swap extents, and - * fs_info->swapfile_pins prevents them from running while the swap file - * is active and moving the extents. Note that this also prevents a - * concurrent device add which isn't actually necessary, but it's not + * under us. The exclop protection makes sure they aren't running/won't + * run concurrently while we are mapping the swap extents, and + * fs_info->swapfile_pins prevents them from running while the swap + * file is active and moving the extents. Note that this also prevents + * a concurrent device add which isn't actually necessary, but it's not * really worth the trouble to allow it. */ - if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) { + if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_SWAP_ACTIVATE)) { btrfs_warn(fs_info, "cannot activate swapfile while exclusive operation is running"); return -EBUSY; } + + /* + * Prevent snapshot creation while we are activating the swap file. + * We do not want to race with snapshot creation. If snapshot creation + * already started before we bumped nr_swapfiles from 0 to 1 and + * completes before the first write into the swap file after it is + * activated, than that write would fallback to COW. + */ + if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) { + btrfs_exclop_finish(fs_info); + btrfs_warn(fs_info, + "cannot activate swapfile because snapshot creation is in progress"); + return -EINVAL; + } /* * Snapshots can create extents which require COW even if NODATACOW is * set. We use this counter to prevent snapshots. We must increment it * before walking the extents because we don't want a concurrent * snapshot to run after we've already checked the extents. + * + * It is possible that subvolume is marked for deletion but still not + * removed yet. To prevent this race, we check the root status before + * activating the swapfile. */ - atomic_inc(&BTRFS_I(inode)->root->nr_swapfiles); + spin_lock(&root->root_item_lock); + if (btrfs_root_dead(root)) { + spin_unlock(&root->root_item_lock); + + btrfs_exclop_finish(fs_info); + btrfs_warn(fs_info, + "cannot activate swapfile because subvolume %llu is being deleted", + root->root_key.objectid); + return -EPERM; + } + atomic_inc(&root->nr_swapfiles); + spin_unlock(&root->root_item_lock); isize = ALIGN_DOWN(inode->i_size, fs_info->sectorsize); - lock_extent_bits(io_tree, 0, isize - 1, &cached_state); + lock_extent(io_tree, 0, isize - 1, &cached_state); start = 0; while (start < isize) { u64 logical_block_start, physical_block_start; @@ -10300,7 +11098,7 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, free_extent_map(em); em = NULL; - ret = can_nocow_extent(inode, start, &len, NULL, NULL, NULL); + ret = can_nocow_extent(inode, start, &len, NULL, NULL, NULL, false, true); if (ret < 0) { goto out; } else if (ret) { @@ -10352,6 +11150,17 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, goto out; } + if (!btrfs_inc_block_group_swap_extents(bg)) { + btrfs_warn(fs_info, + "block group for swapfile at %llu is read-only%s", + bg->start, + atomic_read(&fs_info->scrubs_running) ? + " (scrub running)" : ""); + btrfs_put_block_group(bg); + ret = -EINVAL; + goto out; + } + ret = btrfs_add_swapfile_pin(inode, bg, true); if (ret) { btrfs_put_block_group(bg); @@ -10385,12 +11194,14 @@ out: if (!IS_ERR_OR_NULL(em)) free_extent_map(em); - unlock_extent_cached(io_tree, 0, isize - 1, &cached_state); + unlock_extent(io_tree, 0, isize - 1, &cached_state); if (ret) btrfs_swap_deactivate(file); - clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); + btrfs_drew_write_unlock(&root->snapshot_lock); + + btrfs_exclop_finish(fs_info); if (ret) return ret; @@ -10415,6 +11226,62 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, } #endif +/* + * Update the number of bytes used in the VFS' inode. When we replace extents in + * a range (clone, dedupe, fallocate's zero range), we must update the number of + * bytes used by the inode in an atomic manner, so that concurrent stat(2) calls + * always get a correct value. + */ +void btrfs_update_inode_bytes(struct btrfs_inode *inode, + const u64 add_bytes, + const u64 del_bytes) +{ + if (add_bytes == del_bytes) + return; + + spin_lock(&inode->lock); + if (del_bytes > 0) + inode_sub_bytes(&inode->vfs_inode, del_bytes); + if (add_bytes > 0) + inode_add_bytes(&inode->vfs_inode, add_bytes); + spin_unlock(&inode->lock); +} + +/** + * Verify that there are no ordered extents for a given file range. + * + * @inode: The target inode. + * @start: Start offset of the file range, should be sector size aligned. + * @end: End offset (inclusive) of the file range, its value +1 should be + * sector size aligned. + * + * This should typically be used for cases where we locked an inode's VFS lock in + * exclusive mode, we have also locked the inode's i_mmap_lock in exclusive mode, + * we have flushed all delalloc in the range, we have waited for all ordered + * extents in the range to complete and finally we have locked the file range in + * the inode's io_tree. + */ +void btrfs_assert_inode_range_clean(struct btrfs_inode *inode, u64 start, u64 end) +{ + struct btrfs_root *root = inode->root; + struct btrfs_ordered_extent *ordered; + + if (!IS_ENABLED(CONFIG_BTRFS_ASSERT)) + return; + + ordered = btrfs_lookup_first_ordered_range(inode, start, end + 1 - start); + if (ordered) { + btrfs_err(root->fs_info, +"found unexpected ordered extent in file range [%llu, %llu] for inode %llu root %llu (ordered range [%llu, %llu])", + start, end, btrfs_ino(inode), root->root_key.objectid, + ordered->file_offset, + ordered->file_offset + ordered->num_bytes - 1); + btrfs_put_ordered_extent(ordered); + } + + ASSERT(ordered == NULL); +} + static const struct inode_operations btrfs_dir_inode_operations = { .getattr = btrfs_getattr, .lookup = btrfs_lookup, @@ -10433,6 +11300,8 @@ static const struct inode_operations btrfs_dir_inode_operations = { .set_acl = btrfs_set_acl, .update_time = btrfs_update_time, .tmpfile = btrfs_tmpfile, + .fileattr_get = btrfs_fileattr_get, + .fileattr_set = btrfs_fileattr_set, }; static const struct file_operations btrfs_dir_file_operations = { @@ -10448,12 +11317,6 @@ static const struct file_operations btrfs_dir_file_operations = { .fsync = btrfs_sync_file, }; -static const struct extent_io_ops btrfs_extent_io_ops = { - /* mandatory callbacks */ - .submit_bio_hook = btrfs_submit_bio_hook, - .readpage_end_io_hook = btrfs_readpage_end_io_hook, -}; - /* * btrfs doesn't support the bmap operation because swapfiles * use bmap to make a mapping of extents in the file. They assume @@ -10467,14 +11330,14 @@ static const struct extent_io_ops btrfs_extent_io_ops = { * For now we're avoiding this by dropping bmap. */ static const struct address_space_operations btrfs_aops = { - .readpage = btrfs_readpage, - .writepage = btrfs_writepage, + .read_folio = btrfs_read_folio, .writepages = btrfs_writepages, - .readpages = btrfs_readpages, - .direct_IO = btrfs_direct_IO, - .invalidatepage = btrfs_invalidatepage, - .releasepage = btrfs_releasepage, - .set_page_dirty = btrfs_set_page_dirty, + .readahead = btrfs_readahead, + .direct_IO = noop_direct_IO, + .invalidate_folio = btrfs_invalidate_folio, + .release_folio = btrfs_release_folio, + .migrate_folio = btrfs_migrate_folio, + .dirty_folio = filemap_dirty_folio, .error_remove_page = generic_error_remove_page, .swap_activate = btrfs_swap_activate, .swap_deactivate = btrfs_swap_deactivate, @@ -10489,6 +11352,8 @@ static const struct inode_operations btrfs_file_inode_operations = { .get_acl = btrfs_get_acl, .set_acl = btrfs_set_acl, .update_time = btrfs_update_time, + .fileattr_get = btrfs_fileattr_get, + .fileattr_set = btrfs_fileattr_set, }; static const struct inode_operations btrfs_special_inode_operations = { .getattr = btrfs_getattr, |