From 97d038562a85e64e9929b9cc91f2b35d5f7bf160 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 2 May 2017 12:25:27 +0300 Subject: Btrfs: remove an unused variable "item" is never used. Signed-off-by: Dan Carpenter Signed-off-by: David Sterba --- fs/btrfs/inode.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ef3c98c527c1..ada36206737f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5868,7 +5868,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) struct inode *inode = file_inode(file); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_item *item; struct btrfs_dir_item *di; struct btrfs_key key; struct btrfs_key found_key; @@ -5919,7 +5918,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) continue; } - item = btrfs_item_nr(slot); btrfs_item_key_to_cpu(leaf, &found_key, slot); if (found_key.objectid != key.objectid) -- cgit v1.2.3 From c6100a4b4e3d1650deafda45e49571b83270c714 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 5 May 2017 11:57:13 -0400 Subject: Btrfs: replace tree->mapping with tree->private_data For extent_io tree's we have carried the address_mapping of the inode around in the io tree in order to pull the inode back out for calling into various tree ops hooks. This works fine when everything that has an extent_io_tree has an inode. But we are going to remove the btree_inode, so we need to change this. Instead just have a generic void * for private data that we can initialize with, and have all the tree ops use that instead. This had a lot of cascading changes but should be relatively straightforward. Signed-off-by: Josef Bacik Reviewed-by: Chandan Rajendra Reviewed-by: David Sterba [ minor reordering of the callback prototypes ] Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 1 + fs/btrfs/disk-io.c | 50 ++++++++++++++---------- fs/btrfs/disk-io.h | 6 +-- fs/btrfs/extent_io.c | 52 ++++++++----------------- fs/btrfs/extent_io.h | 21 +++++----- fs/btrfs/inode.c | 82 +++++++++++++++++++++++++++++++--------- fs/btrfs/relocation.c | 3 +- fs/btrfs/tests/extent-io-tests.c | 2 +- fs/btrfs/transaction.c | 2 +- 9 files changed, 128 insertions(+), 91 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index aea1b3cee887..c457cb177340 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3172,6 +3172,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, int btrfs_merge_bio_hook(struct page *page, unsigned long offset, size_t size, struct bio *bio, unsigned long bio_flags); +void btrfs_set_range_writeback(void *private_data, u64 start, u64 end); int btrfs_page_mkwrite(struct vm_fault *vmf); int btrfs_readpage(struct file *file, struct page *page); void btrfs_evict_inode(struct inode *inode); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 9b8df5aaf76e..d8ad30fdeee6 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -118,7 +118,8 @@ void btrfs_end_io_wq_exit(void) * just before they are sent down the IO stack. */ struct async_submit_bio { - struct inode *inode; + void *private_data; + struct btrfs_fs_info *fs_info; struct bio *bio; struct list_head list; extent_submit_bio_hook_t *submit_bio_start; @@ -871,7 +872,7 @@ static void run_one_async_start(struct btrfs_work *work) int ret; async = container_of(work, struct async_submit_bio, work); - ret = async->submit_bio_start(async->inode, async->bio, + ret = async->submit_bio_start(async->private_data, async->bio, async->mirror_num, async->bio_flags, async->bio_offset); if (ret) @@ -885,7 +886,7 @@ static void run_one_async_done(struct btrfs_work *work) int limit; async = container_of(work, struct async_submit_bio, work); - fs_info = BTRFS_I(async->inode)->root->fs_info; + fs_info = async->fs_info; limit = btrfs_async_submit_limit(fs_info); limit = limit * 2 / 3; @@ -904,7 +905,7 @@ static void run_one_async_done(struct btrfs_work *work) return; } - async->submit_bio_done(async->inode, async->bio, async->mirror_num, + async->submit_bio_done(async->private_data, async->bio, async->mirror_num, async->bio_flags, async->bio_offset); } @@ -916,10 +917,9 @@ static void run_one_async_free(struct btrfs_work *work) kfree(async); } -int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, - struct bio *bio, int mirror_num, - unsigned long bio_flags, - u64 bio_offset, +int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, unsigned long bio_flags, + u64 bio_offset, void *private_data, extent_submit_bio_hook_t *submit_bio_start, extent_submit_bio_hook_t *submit_bio_done) { @@ -929,7 +929,8 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, if (!async) return -ENOMEM; - async->inode = inode; + async->private_data = private_data; + async->fs_info = fs_info; async->bio = bio; async->mirror_num = mirror_num; async->submit_bio_start = submit_bio_start; @@ -975,7 +976,7 @@ static int btree_csum_one_bio(struct bio *bio) return ret; } -static int __btree_submit_bio_start(struct inode *inode, struct bio *bio, +static int __btree_submit_bio_start(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { @@ -986,10 +987,11 @@ static int __btree_submit_bio_start(struct inode *inode, struct bio *bio, return btree_csum_one_bio(bio); } -static int __btree_submit_bio_done(struct inode *inode, struct bio *bio, +static int __btree_submit_bio_done(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + struct inode *inode = private_data; int ret; /* @@ -1015,10 +1017,11 @@ static int check_async_write(unsigned long bio_flags) return 1; } -static int btree_submit_bio_hook(struct inode *inode, struct bio *bio, +static int btree_submit_bio_hook(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + struct inode *inode = private_data; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int async = check_async_write(bio_flags); int ret; @@ -1043,8 +1046,8 @@ static int btree_submit_bio_hook(struct inode *inode, struct bio *bio, * kthread helpers are used to submit writes so that * checksumming can happen in parallel across all CPUs */ - ret = btrfs_wq_submit_bio(fs_info, inode, bio, mirror_num, 0, - bio_offset, + ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, 0, + bio_offset, private_data, __btree_submit_bio_start, __btree_submit_bio_done); } @@ -1347,8 +1350,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, root->log_transid_committed = -1; root->last_log_commit = 0; if (!dummy) - extent_io_tree_init(&root->dirty_log_pages, - fs_info->btree_inode->i_mapping); + extent_io_tree_init(&root->dirty_log_pages, NULL); memset(&root->root_key, 0, sizeof(root->root_key)); memset(&root->root_item, 0, sizeof(root->root_item)); @@ -2309,7 +2311,7 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info) inode->i_mapping->a_ops = &btree_aops; RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node); - extent_io_tree_init(&BTRFS_I(inode)->io_tree, inode->i_mapping); + extent_io_tree_init(&BTRFS_I(inode)->io_tree, inode); BTRFS_I(inode)->io_tree.track_uptodate = 0; extent_map_tree_init(&BTRFS_I(inode)->extent_tree); @@ -2703,10 +2705,8 @@ int open_ctree(struct super_block *sb, fs_info->block_group_cache_tree = RB_ROOT; fs_info->first_logical_byte = (u64)-1; - extent_io_tree_init(&fs_info->freed_extents[0], - fs_info->btree_inode->i_mapping); - extent_io_tree_init(&fs_info->freed_extents[1], - fs_info->btree_inode->i_mapping); + extent_io_tree_init(&fs_info->freed_extents[0], NULL); + extent_io_tree_init(&fs_info->freed_extents[1], NULL); fs_info->pinned_extents = &fs_info->freed_extents[0]; set_bit(BTRFS_FS_BARRIER, &fs_info->flags); @@ -4686,6 +4686,12 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) return 0; } +static struct btrfs_fs_info *btree_fs_info(void *private_data) +{ + struct inode *inode = private_data; + return btrfs_sb(inode->i_sb); +} + static const struct extent_io_ops btree_extent_io_ops = { /* mandatory callbacks */ .submit_bio_hook = btree_submit_bio_hook, @@ -4693,6 +4699,8 @@ static const struct extent_io_ops btree_extent_io_ops = { /* note we're sharing with inode.c for the merge bio hook */ .merge_bio_hook = btrfs_merge_bio_hook, .readpage_io_failed_hook = btree_io_failed_hook, + .set_range_writeback = btrfs_set_range_writeback, + .tree_fs_info = btree_fs_info, /* optional callbacks */ }; diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index 21f1ceb85b76..35ddfcf04ad0 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -120,9 +120,9 @@ u32 btrfs_csum_data(const char *data, u32 seed, size_t len); void btrfs_csum_final(u32 crc, u8 *result); int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio, enum btrfs_wq_endio_type metadata); -int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, - struct bio *bio, int mirror_num, - unsigned long bio_flags, u64 bio_offset, +int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, unsigned long bio_flags, + u64 bio_offset, void *private_data, extent_submit_bio_hook_t *submit_bio_start, extent_submit_bio_hook_t *submit_bio_done); unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d3619e010005..70832b9b9e21 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -87,19 +87,9 @@ void btrfs_leak_debug_check(void) static inline void __btrfs_debug_check_extent_io_range(const char *caller, struct extent_io_tree *tree, u64 start, u64 end) { - struct inode *inode; - u64 isize; - - if (!tree->mapping) - return; - - inode = tree->mapping->host; - isize = i_size_read(inode); - if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) { - btrfs_debug_rl(BTRFS_I(inode)->root->fs_info, - "%s: ino %llu isize %llu odd range [%llu,%llu]", - caller, btrfs_ino(BTRFS_I(inode)), isize, start, end); - } + if (tree->ops && tree->ops->check_extent_io_range) + tree->ops->check_extent_io_range(tree->private_data, caller, + start, end); } #else #define btrfs_leak_debug_add(new, head) do {} while (0) @@ -154,9 +144,9 @@ static noinline void flush_write_bio(void *data); static inline struct btrfs_fs_info * tree_fs_info(struct extent_io_tree *tree) { - if (!tree->mapping) - return NULL; - return btrfs_sb(tree->mapping->host->i_sb); + if (tree->ops) + return tree->ops->tree_fs_info(tree->private_data); + return NULL; } int __init extent_io_init(void) @@ -213,13 +203,13 @@ void extent_io_exit(void) } void extent_io_tree_init(struct extent_io_tree *tree, - struct address_space *mapping) + void *private_data) { tree->state = RB_ROOT; tree->ops = NULL; tree->dirty_bytes = 0; spin_lock_init(&tree->lock); - tree->mapping = mapping; + tree->private_data = private_data; } static struct extent_state *alloc_extent_state(gfp_t mask) @@ -369,8 +359,7 @@ static void merge_cb(struct extent_io_tree *tree, struct extent_state *new, struct extent_state *other) { if (tree->ops && tree->ops->merge_extent_hook) - tree->ops->merge_extent_hook(tree->mapping->host, new, - other); + tree->ops->merge_extent_hook(tree->private_data, new, other); } /* @@ -421,15 +410,14 @@ static void set_state_cb(struct extent_io_tree *tree, struct extent_state *state, unsigned *bits) { if (tree->ops && tree->ops->set_bit_hook) - tree->ops->set_bit_hook(tree->mapping->host, state, bits); + tree->ops->set_bit_hook(tree->private_data, state, bits); } static void clear_state_cb(struct extent_io_tree *tree, struct extent_state *state, unsigned *bits) { if (tree->ops && tree->ops->clear_bit_hook) - tree->ops->clear_bit_hook(BTRFS_I(tree->mapping->host), - state, bits); + tree->ops->clear_bit_hook(tree->private_data, state, bits); } static void set_state_bits(struct extent_io_tree *tree, @@ -478,7 +466,7 @@ static void split_cb(struct extent_io_tree *tree, struct extent_state *orig, u64 split) { if (tree->ops && tree->ops->split_extent_hook) - tree->ops->split_extent_hook(tree->mapping->host, orig, split); + tree->ops->split_extent_hook(tree->private_data, orig, split); } /* @@ -1402,17 +1390,7 @@ void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end) */ static void set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end) { - unsigned long index = start >> PAGE_SHIFT; - unsigned long end_index = end >> PAGE_SHIFT; - struct page *page; - - while (index <= end_index) { - page = find_get_page(tree->mapping, index); - BUG_ON(!page); /* Pages should be in the extent_io_tree */ - set_page_writeback(page); - put_page(page); - index++; - } + tree->ops->set_range_writeback(tree->private_data, start, end); } /* find the first state struct with 'bits' set after 'start', and @@ -2431,7 +2409,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, "Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d", read_mode, failrec->this_mirror, failrec->in_validation); - ret = tree->ops->submit_bio_hook(inode, bio, failrec->this_mirror, + ret = tree->ops->submit_bio_hook(tree->private_data, bio, failrec->this_mirror, failrec->bio_flags, 0); if (ret) { free_io_failure(BTRFS_I(inode), failrec); @@ -2755,7 +2733,7 @@ static int __must_check submit_one_bio(struct bio *bio, int mirror_num, bio_get(bio); if (tree->ops) - ret = tree->ops->submit_bio_hook(page->mapping->host, bio, + ret = tree->ops->submit_bio_hook(tree->private_data, bio, mirror_num, bio_flags, start); else btrfsic_submit_bio(bio); diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 1eafa2f0ede3..15ef696dda51 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -92,7 +92,7 @@ struct btrfs_inode; struct btrfs_io_bio; struct io_failure_record; -typedef int (extent_submit_bio_hook_t)(struct inode *inode, struct bio *bio, +typedef int (extent_submit_bio_hook_t)(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset); struct extent_io_ops { @@ -108,32 +108,36 @@ struct extent_io_ops { size_t size, struct bio *bio, unsigned long bio_flags); int (*readpage_io_failed_hook)(struct page *page, int failed_mirror); + struct btrfs_fs_info *(*tree_fs_info)(void *private_data); + void (*set_range_writeback)(void *private_data, u64 start, u64 end); /* * Optional hooks, called if the pointer is not NULL */ - int (*fill_delalloc)(struct inode *inode, struct page *locked_page, + int (*fill_delalloc)(void *private_data, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written); int (*writepage_start_hook)(struct page *page, u64 start, u64 end); void (*writepage_end_io_hook)(struct page *page, u64 start, u64 end, struct extent_state *state, int uptodate); - void (*set_bit_hook)(struct inode *inode, struct extent_state *state, + void (*set_bit_hook)(void *private_data, struct extent_state *state, unsigned *bits); - void (*clear_bit_hook)(struct btrfs_inode *inode, + void (*clear_bit_hook)(void *private_data, struct extent_state *state, unsigned *bits); - void (*merge_extent_hook)(struct inode *inode, + void (*merge_extent_hook)(void *private_data, struct extent_state *new, struct extent_state *other); - void (*split_extent_hook)(struct inode *inode, + void (*split_extent_hook)(void *private_data, struct extent_state *orig, u64 split); + void (*check_extent_io_range)(void *private_data, const char *caller, + u64 start, u64 end); }; struct extent_io_tree { struct rb_root state; - struct address_space *mapping; + void *private_data; u64 dirty_bytes; int track_uptodate; spinlock_t lock; @@ -230,8 +234,7 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode, u64 start, u64 len, int create); -void extent_io_tree_init(struct extent_io_tree *tree, - struct address_space *mapping); +void extent_io_tree_init(struct extent_io_tree *tree, void *private_data); int try_release_extent_mapping(struct extent_map_tree *map, struct extent_io_tree *tree, struct page *page, gfp_t mask); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ada36206737f..54aa757d4cc3 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1569,10 +1569,11 @@ static inline int need_force_cow(struct inode *inode, u64 start, u64 end) /* * extent_io.c call back to do delayed allocation processing */ -static int run_delalloc_range(struct inode *inode, struct page *locked_page, +static int run_delalloc_range(void *private_data, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written) { + struct inode *inode = private_data; int ret; int force_cow = need_force_cow(inode, start, end); @@ -1596,9 +1597,10 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page, return ret; } -static void btrfs_split_extent_hook(struct inode *inode, +static void btrfs_split_extent_hook(void *private_data, struct extent_state *orig, u64 split) { + struct inode *inode = private_data; u64 size; /* not delalloc, ignore it */ @@ -1633,10 +1635,11 @@ static void btrfs_split_extent_hook(struct inode *inode, * extents, such as when we are doing sequential writes, so we can properly * account for the metadata space we'll need. */ -static void btrfs_merge_extent_hook(struct inode *inode, +static void btrfs_merge_extent_hook(void *private_data, struct extent_state *new, struct extent_state *other) { + struct inode *inode = private_data; u64 new_size, old_size; u32 num_extents; @@ -1736,9 +1739,10 @@ static void btrfs_del_delalloc_inode(struct btrfs_root *root, * bytes in this file, and to maintain the list of inodes that * have pending delalloc work to be done. */ -static void btrfs_set_bit_hook(struct inode *inode, +static void btrfs_set_bit_hook(void *private_data, struct extent_state *state, unsigned *bits) { + struct inode *inode = private_data; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); @@ -1790,10 +1794,11 @@ static void btrfs_set_bit_hook(struct inode *inode, /* * extent_io.c clear_bit_hook, see set_bit_hook for why */ -static void btrfs_clear_bit_hook(struct btrfs_inode *inode, +static void btrfs_clear_bit_hook(void *private_data, struct extent_state *state, unsigned *bits) { + struct btrfs_inode *inode = BTRFS_I((struct inode *)private_data); struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); u64 len = state->end + 1 - state->start; u32 num_extents = count_max_extents(len); @@ -1901,10 +1906,11 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset, * At IO completion time the cums attached on the ordered extent record * are inserted into the btree */ -static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio, +static int __btrfs_submit_bio_start(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + struct inode *inode = private_data; int ret = 0; ret = btrfs_csum_one_bio(inode, bio, 0, 0); @@ -1920,10 +1926,11 @@ static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio, * At IO completion time the cums attached on the ordered extent record * are inserted into the btree */ -static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio, +static int __btrfs_submit_bio_done(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + struct inode *inode = private_data; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int ret; @@ -1939,10 +1946,11 @@ static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio, * 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 */ -static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, - int mirror_num, unsigned long bio_flags, - u64 bio_offset) +static int btrfs_submit_bio_hook(void *private_data, struct bio *bio, + int mirror_num, unsigned long bio_flags, + u64 bio_offset) { + struct inode *inode = private_data; 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; @@ -1976,8 +1984,8 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio, if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID) goto mapit; /* we're doing a write, do the async checksumming */ - ret = btrfs_wq_submit_bio(fs_info, inode, bio, mirror_num, - bio_flags, bio_offset, + ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, bio_flags, + bio_offset, inode, __btrfs_submit_bio_start, __btrfs_submit_bio_done); goto out; @@ -8306,10 +8314,11 @@ static void btrfs_endio_direct_write(struct bio *bio) bio_put(bio); } -static int __btrfs_submit_bio_start_direct_io(struct inode *inode, +static int __btrfs_submit_bio_start_direct_io(void *private_data, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 offset) { + struct inode *inode = private_data; int ret; ret = btrfs_csum_one_bio(inode, bio, offset, 1); BUG_ON(ret); /* -ENOMEM */ @@ -8421,8 +8430,8 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, goto map; if (write && async_submit) { - ret = btrfs_wq_submit_bio(fs_info, inode, bio, 0, 0, - file_offset, + ret = btrfs_wq_submit_bio(fs_info, bio, 0, 0, + file_offset, inode, __btrfs_submit_bio_start_direct_io, __btrfs_submit_bio_done); goto err; @@ -9402,8 +9411,8 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) inode = &ei->vfs_inode; extent_map_tree_init(&ei->extent_tree); - extent_io_tree_init(&ei->io_tree, &inode->i_data); - extent_io_tree_init(&ei->io_failure_tree, &inode->i_data); + extent_io_tree_init(&ei->io_tree, inode); + extent_io_tree_init(&ei->io_failure_tree, inode); ei->io_tree.track_uptodate = 1; ei->io_failure_tree.track_uptodate = 1; atomic_set(&ei->sync_writers, 0); @@ -10657,6 +10666,42 @@ static int btrfs_readpage_io_failed_hook(struct page *page, int failed_mirror) return -EAGAIN; } +static struct btrfs_fs_info *iotree_fs_info(void *private_data) +{ + struct inode *inode = private_data; + return btrfs_sb(inode->i_sb); +} + +static void btrfs_check_extent_io_range(void *private_data, const char *caller, + u64 start, u64 end) +{ + struct inode *inode = private_data; + u64 isize; + + isize = i_size_read(inode); + if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) { + btrfs_debug_rl(BTRFS_I(inode)->root->fs_info, + "%s: ino %llu isize %llu odd range [%llu,%llu]", + caller, btrfs_ino(BTRFS_I(inode)), isize, start, end); + } +} + +void btrfs_set_range_writeback(void *private_data, u64 start, u64 end) +{ + struct inode *inode = private_data; + unsigned long index = start >> PAGE_SHIFT; + unsigned long end_index = end >> PAGE_SHIFT; + struct page *page; + + while (index <= end_index) { + page = find_get_page(inode->i_mapping, index); + ASSERT(page); /* Pages should be in the extent_io_tree */ + set_page_writeback(page); + put_page(page); + index++; + } +} + static const struct inode_operations btrfs_dir_inode_operations = { .getattr = btrfs_getattr, .lookup = btrfs_lookup, @@ -10700,6 +10745,8 @@ static const struct extent_io_ops btrfs_extent_io_ops = { .readpage_end_io_hook = btrfs_readpage_end_io_hook, .merge_bio_hook = btrfs_merge_bio_hook, .readpage_io_failed_hook = btrfs_readpage_io_failed_hook, + .tree_fs_info = iotree_fs_info, + .set_range_writeback = btrfs_set_range_writeback, /* optional callbacks */ .fill_delalloc = run_delalloc_range, @@ -10709,6 +10756,7 @@ static const struct extent_io_ops btrfs_extent_io_ops = { .clear_bit_hook = btrfs_clear_bit_hook, .merge_extent_hook = btrfs_merge_extent_hook, .split_extent_hook = btrfs_split_extent_hook, + .check_extent_io_range = btrfs_check_extent_io_range, }; /* diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index d60df51959f7..b291d1bebb4c 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4269,8 +4269,7 @@ static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info) INIT_LIST_HEAD(&rc->reloc_roots); backref_cache_init(&rc->backref_cache); mapping_tree_init(&rc->reloc_root_tree); - extent_io_tree_init(&rc->processed_blocks, - fs_info->btree_inode->i_mapping); + extent_io_tree_init(&rc->processed_blocks, NULL); return rc; } diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c index 133753232a94..d06b1c931d05 100644 --- a/fs/btrfs/tests/extent-io-tests.c +++ b/fs/btrfs/tests/extent-io-tests.c @@ -87,7 +87,7 @@ static int test_find_delalloc(u32 sectorsize) return -ENOMEM; } - extent_io_tree_init(&tmp, &inode->i_data); + extent_io_tree_init(&tmp, inode); /* * First go through and create and mark all of our pages dirty, we pin diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 2168654c90a1..ca0009ff47f1 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -294,7 +294,7 @@ loop: spin_lock_init(&cur_trans->dropped_roots_lock); list_add_tail(&cur_trans->list, &fs_info->trans_list); extent_io_tree_init(&cur_trans->dirty_pages, - fs_info->btree_inode->i_mapping); + fs_info->btree_inode); fs_info->generation++; cur_trans->transid = fs_info->generation; fs_info->running_transaction = cur_trans; -- cgit v1.2.3 From 7870d0822be99bdb9353b542007c046966ec18f3 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 5 May 2017 11:57:15 -0400 Subject: Btrfs: don't pass the inode through clean_io_failure Instead pass around the failure tree and the io tree. Signed-off-by: Josef Bacik Reviewed-by: Chandan Rajendra Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 51 ++++++++++++++++++++++++++++----------------------- fs/btrfs/extent_io.h | 10 +++++++--- fs/btrfs/inode.c | 37 ++++++++++++++++++++++++++----------- 3 files changed, 61 insertions(+), 37 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 47ebaec85096..1fcfa1d9e77a 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1939,11 +1939,12 @@ static void check_page_uptodate(struct extent_io_tree *tree, struct page *page) SetPageUptodate(page); } -int free_io_failure(struct btrfs_inode *inode, struct io_failure_record *rec) +int free_io_failure(struct extent_io_tree *failure_tree, + struct extent_io_tree *io_tree, + struct io_failure_record *rec) { int ret; int err = 0; - struct extent_io_tree *failure_tree = &inode->io_failure_tree; set_state_failrec(failure_tree, rec->start, NULL); ret = clear_extent_bits(failure_tree, rec->start, @@ -1952,7 +1953,7 @@ int free_io_failure(struct btrfs_inode *inode, struct io_failure_record *rec) if (ret) err = ret; - ret = clear_extent_bits(&inode->io_tree, rec->start, + ret = clear_extent_bits(io_tree, rec->start, rec->start + rec->len - 1, EXTENT_DAMAGED); if (ret && !err) @@ -2081,24 +2082,24 @@ int repair_eb_io_failure(struct btrfs_fs_info *fs_info, * each time an IO finishes, we do a fast check in the IO failure tree * to see if we need to process or clean up an io_failure_record */ -int clean_io_failure(struct btrfs_inode *inode, u64 start, struct page *page, - unsigned int pg_offset) +int clean_io_failure(struct btrfs_fs_info *fs_info, + struct extent_io_tree *failure_tree, + struct extent_io_tree *io_tree, u64 start, + struct page *page, u64 ino, unsigned int pg_offset) { u64 private; struct io_failure_record *failrec; - struct btrfs_fs_info *fs_info = inode->root->fs_info; struct extent_state *state; int num_copies; int ret; private = 0; - ret = count_range_bits(&inode->io_failure_tree, &private, - (u64)-1, 1, EXTENT_DIRTY, 0); + ret = count_range_bits(failure_tree, &private, (u64)-1, 1, + EXTENT_DIRTY, 0); if (!ret) return 0; - ret = get_state_failrec(&inode->io_failure_tree, start, - &failrec); + ret = get_state_failrec(failure_tree, start, &failrec); if (ret) return 0; @@ -2114,25 +2115,25 @@ int clean_io_failure(struct btrfs_inode *inode, u64 start, struct page *page, if (fs_info->sb->s_flags & MS_RDONLY) goto out; - spin_lock(&inode->io_tree.lock); - state = find_first_extent_bit_state(&inode->io_tree, + spin_lock(&io_tree->lock); + state = find_first_extent_bit_state(io_tree, failrec->start, EXTENT_LOCKED); - spin_unlock(&inode->io_tree.lock); + spin_unlock(&io_tree->lock); if (state && state->start <= failrec->start && state->end >= failrec->start + failrec->len - 1) { num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len); if (num_copies > 1) { - repair_io_failure(fs_info, btrfs_ino(inode), start, - failrec->len, failrec->logical, page, - pg_offset, failrec->failed_mirror); + repair_io_failure(fs_info, ino, start, failrec->len, + failrec->logical, page, pg_offset, + failrec->failed_mirror); } } out: - free_io_failure(inode, failrec); + free_io_failure(failure_tree, io_tree, failrec); return 0; } @@ -2373,6 +2374,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, struct io_failure_record *failrec; struct inode *inode = page->mapping->host; struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; + struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree; struct bio *bio; int read_mode = 0; int ret; @@ -2385,7 +2387,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, ret = btrfs_check_repairable(inode, failed_bio, failrec, failed_mirror); if (!ret) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, tree, failrec); return -EIO; } @@ -2398,7 +2400,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, (int)phy_offset, failed_bio->bi_end_io, NULL); if (!bio) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, tree, failrec); return -EIO; } bio_set_op_attrs(bio, REQ_OP_READ, read_mode); @@ -2410,7 +2412,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, ret = tree->ops->submit_bio_hook(tree->private_data, bio, failrec->this_mirror, failrec->bio_flags, 0); if (ret) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, tree, failrec); bio_put(bio); } @@ -2514,7 +2516,7 @@ static void end_bio_extent_readpage(struct bio *bio) struct bio_vec *bvec; int uptodate = !bio->bi_error; struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); - struct extent_io_tree *tree; + struct extent_io_tree *tree, *failure_tree; u64 offset = 0; u64 start; u64 end; @@ -2535,6 +2537,7 @@ static void end_bio_extent_readpage(struct bio *bio) (u64)bio->bi_iter.bi_sector, bio->bi_error, io_bio->mirror_num); tree = &BTRFS_I(inode)->io_tree; + failure_tree = &BTRFS_I(inode)->io_failure_tree; /* We always issue full-page reads, but if some block * in a page fails to read, blk_update_request() will @@ -2564,8 +2567,10 @@ static void end_bio_extent_readpage(struct bio *bio) if (ret) uptodate = 0; else - clean_io_failure(BTRFS_I(inode), start, - page, 0); + clean_io_failure(BTRFS_I(inode)->root->fs_info, + failure_tree, tree, start, + page, + btrfs_ino(BTRFS_I(inode)), 0); } if (likely(uptodate)) diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index aa3b1fcfc15f..1383afdf1eeb 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -474,8 +474,10 @@ struct btrfs_inode; int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start, u64 length, u64 logical, struct page *page, unsigned int pg_offset, int mirror_num); -int clean_io_failure(struct btrfs_inode *inode, u64 start, - struct page *page, unsigned int pg_offset); +int clean_io_failure(struct btrfs_fs_info *fs_info, + struct extent_io_tree *failure_tree, + struct extent_io_tree *io_tree, u64 start, + struct page *page, u64 ino, unsigned int pg_offset); void end_extent_writepage(struct page *page, int err, u64 start, u64 end); int repair_eb_io_failure(struct btrfs_fs_info *fs_info, struct extent_buffer *eb, int mirror_num); @@ -510,7 +512,9 @@ struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio, struct io_failure_record *failrec, struct page *page, int pg_offset, int icsum, bio_end_io_t *endio_func, void *data); -int free_io_failure(struct btrfs_inode *inode, struct io_failure_record *rec); +int free_io_failure(struct extent_io_tree *failure_tree, + struct extent_io_tree *io_tree, + struct io_failure_record *rec); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS noinline u64 find_lock_delalloc_range(struct inode *inode, struct extent_io_tree *tree, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 54aa757d4cc3..c61881df2338 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7984,6 +7984,8 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, 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; int read_mode = 0; @@ -7998,7 +8000,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, ret = btrfs_check_dio_repairable(inode, failed_bio, failrec, failed_mirror); if (!ret) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, io_tree, failrec); return -EIO; } @@ -8012,7 +8014,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page, pgoff, isector, repair_endio, repair_arg); if (!bio) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, io_tree, failrec); return -EIO; } bio_set_op_attrs(bio, REQ_OP_READ, read_mode); @@ -8023,7 +8025,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror); if (ret) { - free_io_failure(BTRFS_I(inode), failrec); + free_io_failure(failure_tree, io_tree, failrec); bio_put(bio); } @@ -8040,19 +8042,24 @@ struct btrfs_retry_complete { 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; int i; if (bio->bi_error) goto end; ASSERT(bio->bi_vcnt == 1); - ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode)); + io_tree = &BTRFS_I(inode)->io_tree; + failure_tree = &BTRFS_I(inode)->io_failure_tree; + ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(inode)); done->uptodate = 1; bio_for_each_segment_all(bvec, bio, i) - clean_io_failure(BTRFS_I(done->inode), done->start, - bvec->bv_page, 0); + 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); @@ -8117,6 +8124,8 @@ static void btrfs_retry_endio(struct bio *bio) { struct btrfs_retry_complete *done = bio->bi_private; struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); + struct extent_io_tree *io_tree, *failure_tree; + struct inode *inode = done->inode; struct bio_vec *bvec; int uptodate; int ret; @@ -8130,13 +8139,19 @@ static void btrfs_retry_endio(struct bio *bio) ASSERT(bio->bi_vcnt == 1); ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode)); + io_tree = &BTRFS_I(inode)->io_tree; + failure_tree = &BTRFS_I(inode)->io_failure_tree; + bio_for_each_segment_all(bvec, bio, i) { - ret = __readpage_endio_check(done->inode, io_bio, i, - bvec->bv_page, bvec->bv_offset, - done->start, bvec->bv_len); + 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(done->inode), done->start, - bvec->bv_page, bvec->bv_offset); + 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; } -- cgit v1.2.3 From 725130bac5857ac035cf9f47d2973ee4e734db54 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Tue, 16 May 2017 09:51:39 -0700 Subject: Btrfs: use bio_clone_bioset_partial to simplify DIO submit Currently when mapping bio to limit bio to a single stripe length, we split bio by adding page to bio one by one, but later we don't modify the vector of bio at all, thus we can use bio_clone_fast to use the original bio vector directly. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 119 +++++++++++++++++++++---------------------------------- 1 file changed, 45 insertions(+), 74 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c61881df2338..c2e9c51c650e 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8380,16 +8380,6 @@ out: bio_put(bio); } -static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev, - u64 first_sector, gfp_t gfp_flags) -{ - struct bio *bio; - bio = btrfs_bio_alloc(bdev, first_sector, BIO_MAX_PAGES, gfp_flags); - if (bio) - bio_associate_current(bio); - return bio; -} - static inline int btrfs_lookup_and_bind_dio_csum(struct inode *inode, struct btrfs_dio_private *dip, struct bio *bio, @@ -8479,24 +8469,23 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, struct btrfs_root *root = BTRFS_I(inode)->root; struct bio *bio; struct bio *orig_bio = dip->orig_bio; - struct bio_vec *bvec; u64 start_sector = orig_bio->bi_iter.bi_sector; u64 file_offset = dip->logical_offset; - u64 submit_len = 0; u64 map_length; - u32 blocksize = fs_info->sectorsize; int async_submit = 0; - int nr_sectors; + u64 submit_len; + int clone_offset = 0; + int clone_len; int ret; - int i, j; map_length = orig_bio->bi_iter.bi_size; + submit_len = map_length; ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), start_sector << 9, &map_length, NULL, 0); if (ret) return -EIO; - if (map_length >= orig_bio->bi_iter.bi_size) { + if (map_length >= submit_len) { bio = orig_bio; dip->flags |= BTRFS_DIO_ORIG_BIO_SUBMITTED; goto submit; @@ -8508,70 +8497,52 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, else async_submit = 1; - bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS); - if (!bio) - return -ENOMEM; - - bio->bi_opf = orig_bio->bi_opf; - bio->bi_private = dip; - bio->bi_end_io = btrfs_end_dio_bio; - btrfs_io_bio(bio)->logical = file_offset; + /* bio split */ + ASSERT(map_length <= INT_MAX); atomic_inc(&dip->pending_bios); + while (submit_len > 0) { + clone_len = min_t(int, submit_len, map_length); - bio_for_each_segment_all(bvec, orig_bio, j) { - nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len); - i = 0; -next_block: - if (unlikely(map_length < submit_len + blocksize || - bio_add_page(bio, bvec->bv_page, blocksize, - bvec->bv_offset + (i * blocksize)) < blocksize)) { - /* - * inc the count before we submit the bio so - * we know the end IO handler won't happen before - * we inc the count. Otherwise, the dip might get freed - * before we're done setting it up - */ - atomic_inc(&dip->pending_bios); - ret = __btrfs_submit_dio_bio(bio, inode, - file_offset, skip_sum, - async_submit); - if (ret) { - bio_put(bio); - atomic_dec(&dip->pending_bios); - goto out_err; - } - - start_sector += submit_len >> 9; - file_offset += submit_len; + /* + * 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, GFP_NOFS, clone_offset, + clone_len); + bio->bi_private = dip; + bio->bi_end_io = btrfs_end_dio_bio; + btrfs_io_bio(bio)->logical = file_offset; + + ASSERT(submit_len >= clone_len); + submit_len -= clone_len; + if (submit_len == 0) + break; - submit_len = 0; + /* + * 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. + */ + atomic_inc(&dip->pending_bios); - bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, - start_sector, GFP_NOFS); - if (!bio) - goto out_err; - bio->bi_opf = orig_bio->bi_opf; - bio->bi_private = dip; - bio->bi_end_io = btrfs_end_dio_bio; - btrfs_io_bio(bio)->logical = file_offset; + ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, + async_submit); + if (ret) { + bio_put(bio); + atomic_dec(&dip->pending_bios); + goto out_err; + } - map_length = orig_bio->bi_iter.bi_size; - ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), - start_sector << 9, - &map_length, NULL, 0); - if (ret) { - bio_put(bio); - goto out_err; - } + clone_offset += clone_len; + start_sector += clone_len >> 9; + file_offset += clone_len; - goto next_block; - } else { - submit_len += blocksize; - if (--nr_sectors) { - i++; - goto next_block; - } - } + map_length = submit_len; + ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), + start_sector << 9, &map_length, NULL, 0); + if (ret) + goto out_err; } submit: -- cgit v1.2.3 From 17347cec15f919901c90cdbf98af47ed2ed28b9f Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 15 May 2017 15:33:27 -0700 Subject: Btrfs: change how we iterate bios in endio Since dio submit has used bio_clone_fast, the submitted bio may not have a reliable bi_vcnt, for the bio vector iterations in checksum related functions, bio->bi_iter is not modified yet and it's safe to use bio_for_each_segment, while for those bio vector iterations in dio read's endio, we now save a copy of bvec_iter in struct btrfs_io_bio when cloning bios and use the helper __bio_for_each_segment with the saved bvec_iter to access each bvec. Also for dio reads which don't get split, we also need to save a copy of bio iterator in btrfs_bio_clone to let __bio_for_each_segments to access each bvec in dio read's endio. Note that it doesn't affect other calls of btrfs_bio_clone() because they don't need to use this iterator. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 2 ++ fs/btrfs/file-item.c | 31 +++++++++++++++---------------- fs/btrfs/inode.c | 35 +++++++++++++++++++---------------- fs/btrfs/volumes.h | 1 + 4 files changed, 37 insertions(+), 32 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index db2bbf92b4bc..acb8c1d177e2 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2700,6 +2700,7 @@ struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask) btrfs_bio->csum = NULL; btrfs_bio->csum_allocated = NULL; btrfs_bio->end_io = NULL; + btrfs_bio->iter = bio->bi_iter; } return new; } @@ -2736,6 +2737,7 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, gfp_t gfp_mask, btrfs_bio->end_io = NULL; bio_trim(bio, offset >> 9, size >> 9); + btrfs_bio->iter = bio->bi_iter; return bio; } diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 64fcb31d7163..9f6062c82419 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -164,7 +164,8 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u64 logical_offset, u32 *dst, int dio) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct bio_vec *bvec; + struct bio_vec bvec; + struct bvec_iter iter; struct btrfs_io_bio *btrfs_bio = btrfs_io_bio(bio); struct btrfs_csum_item *item = NULL; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; @@ -177,7 +178,7 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u64 page_bytes_left; u32 diff; int nblocks; - int count = 0, i; + int count = 0; u16 csum_size = btrfs_super_csum_size(fs_info->super_copy); path = btrfs_alloc_path(); @@ -206,8 +207,6 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, if (bio->bi_iter.bi_size > PAGE_SIZE * 8) path->reada = READA_FORWARD; - WARN_ON(bio->bi_vcnt <= 0); - /* * the free space stuff is only read when it hasn't been * updated in the current transaction. So, we can safely @@ -223,13 +222,13 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, if (dio) offset = logical_offset; - bio_for_each_segment_all(bvec, bio, i) { - page_bytes_left = bvec->bv_len; + bio_for_each_segment(bvec, bio, iter) { + page_bytes_left = bvec.bv_len; if (count) goto next; if (!dio) - offset = page_offset(bvec->bv_page) + bvec->bv_offset; + offset = page_offset(bvec.bv_page) + bvec.bv_offset; count = btrfs_find_ordered_sum(inode, offset, disk_bytenr, (u32 *)csum, nblocks); if (count) @@ -440,15 +439,15 @@ int btrfs_csum_one_bio(struct inode *inode, struct bio *bio, struct btrfs_ordered_sum *sums; struct btrfs_ordered_extent *ordered = NULL; char *data; - struct bio_vec *bvec; + struct bvec_iter iter; + struct bio_vec bvec; int index; int nr_sectors; - int i, j; unsigned long total_bytes = 0; unsigned long this_sum_bytes = 0; + int i; u64 offset; - WARN_ON(bio->bi_vcnt <= 0); sums = kzalloc(btrfs_ordered_sum_size(fs_info, bio->bi_iter.bi_size), GFP_NOFS); if (!sums) @@ -465,19 +464,19 @@ int btrfs_csum_one_bio(struct inode *inode, struct bio *bio, sums->bytenr = (u64)bio->bi_iter.bi_sector << 9; index = 0; - bio_for_each_segment_all(bvec, bio, j) { + bio_for_each_segment(bvec, bio, iter) { if (!contig) - offset = page_offset(bvec->bv_page) + bvec->bv_offset; + offset = page_offset(bvec.bv_page) + bvec.bv_offset; if (!ordered) { ordered = btrfs_lookup_ordered_extent(inode, offset); BUG_ON(!ordered); /* Logic error */ } - data = kmap_atomic(bvec->bv_page); + data = kmap_atomic(bvec.bv_page); nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, - bvec->bv_len + fs_info->sectorsize + bvec.bv_len + fs_info->sectorsize - 1); for (i = 0; i < nr_sectors; i++) { @@ -504,12 +503,12 @@ int btrfs_csum_one_bio(struct inode *inode, struct bio *bio, + total_bytes; index = 0; - data = kmap_atomic(bvec->bv_page); + data = kmap_atomic(bvec.bv_page); } sums->sums[index] = ~(u32)0; sums->sums[index] - = btrfs_csum_data(data + bvec->bv_offset + = btrfs_csum_data(data + bvec.bv_offset + (i * fs_info->sectorsize), sums->sums[index], fs_info->sectorsize); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c2e9c51c650e..ce6a9b2cce51 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7989,6 +7989,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, struct bio *bio; int isector; int read_mode = 0; + int segs; int ret; BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE); @@ -8004,9 +8005,9 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, return -EIO; } - if ((failed_bio->bi_vcnt > 1) - || (failed_bio->bi_io_vec->bv_len - > btrfs_inode_sectorsize(inode))) + segs = bio_segments(failed_bio); + if (segs > 1 || + (failed_bio->bi_io_vec->bv_len > btrfs_inode_sectorsize(inode))) read_mode |= REQ_FAILFAST_DEV; isector = start - btrfs_io_bio(failed_bio)->logical; @@ -8069,13 +8070,13 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, struct btrfs_io_bio *io_bio) { struct btrfs_fs_info *fs_info; - struct bio_vec *bvec; + struct bio_vec bvec; + struct bvec_iter iter; struct btrfs_retry_complete done; u64 start; unsigned int pgoff; u32 sectorsize; int nr_sectors; - int i; int ret; fs_info = BTRFS_I(inode)->root->fs_info; @@ -8083,17 +8084,18 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, start = io_bio->logical; done.inode = inode; + io_bio->bio.bi_iter = io_bio->iter; - bio_for_each_segment_all(bvec, &io_bio->bio, i) { - nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len); - pgoff = bvec->bv_offset; + 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, + 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); @@ -8166,7 +8168,8 @@ static int __btrfs_subio_endio_read(struct inode *inode, struct btrfs_io_bio *io_bio, int err) { struct btrfs_fs_info *fs_info; - struct bio_vec *bvec; + struct bio_vec bvec; + struct bvec_iter iter; struct btrfs_retry_complete done; u64 start; u64 offset = 0; @@ -8174,7 +8177,6 @@ static int __btrfs_subio_endio_read(struct inode *inode, int nr_sectors; unsigned int pgoff; int csum_pos; - int i; int ret; fs_info = BTRFS_I(inode)->root->fs_info; @@ -8183,15 +8185,16 @@ static int __btrfs_subio_endio_read(struct inode *inode, err = 0; start = io_bio->logical; done.inode = inode; + io_bio->bio.bi_iter = io_bio->iter; - bio_for_each_segment_all(bvec, &io_bio->bio, i) { - nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len); + bio_for_each_segment(bvec, &io_bio->bio, iter) { + nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len); - pgoff = bvec->bv_offset; + pgoff = bvec.bv_offset; next_block: csum_pos = BTRFS_BYTES_TO_BLKS(fs_info, offset); ret = __readpage_endio_check(inode, io_bio, csum_pos, - bvec->bv_page, pgoff, start, + bvec.bv_page, pgoff, start, sectorsize); if (likely(!ret)) goto next; @@ -8200,7 +8203,7 @@ try_again: done.start = start; init_completion(&done.done); - ret = dio_read_error(inode, &io_bio->bio, bvec->bv_page, + ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page, pgoff, start, start + sectorsize - 1, io_bio->mirror_num, btrfs_retry_endio, &done); diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 094521729df1..58b97b6f5f02 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -280,6 +280,7 @@ struct btrfs_io_bio { u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE]; u8 *csum_allocated; btrfs_io_bio_end_io_t *end_io; + struct bvec_iter iter; struct bio bio; }; -- cgit v1.2.3 From 629ebf4fadbc647e2118ef9c94b038eeef178e9c Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 15 May 2017 17:20:07 -0700 Subject: Btrfs: record error if one block has failed to retry In the nocsum case of dio read endio, it returns immediately if an error gets returned when repairing, which leaves the rest blocks unrepaired. The behavior is different from how buffered read endio works in the same case. This changes it to record error only and go on repairing the rest blocks. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ce6a9b2cce51..d8d1fe5096b7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8078,6 +8078,7 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, u32 sectorsize; int nr_sectors; int ret; + int err = 0; fs_info = BTRFS_I(inode)->root->fs_info; sectorsize = fs_info->sectorsize; @@ -8099,8 +8100,10 @@ next_block_or_try_again: pgoff, start, start + sectorsize - 1, io_bio->mirror_num, btrfs_retry_endio_nocsum, &done); - if (ret) - return ret; + if (ret) { + err = ret; + goto next; + } wait_for_completion(&done.done); @@ -8109,6 +8112,7 @@ next_block_or_try_again: goto next_block_or_try_again; } +next: start += sectorsize; nr_sectors--; @@ -8119,7 +8123,7 @@ next_block_or_try_again: } } - return 0; + return err; } static void btrfs_retry_endio(struct bio *bio) -- cgit v1.2.3 From 3892ac9086a82c29a6ce71f8b54f820129947119 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 17 Apr 2017 15:00:28 -0700 Subject: Btrfs: unify naming of btrfs_io_bio All dio endio functions are using io_bio for struct btrfs_io_bio, this makes btrfs_submit_direct to follow this convention. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d8d1fe5096b7..53f2cf4b24b6 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8577,16 +8577,16 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode, loff_t file_offset) { struct btrfs_dio_private *dip = NULL; - struct bio *io_bio = NULL; - struct btrfs_io_bio *btrfs_bio; + struct bio *bio = NULL; + struct btrfs_io_bio *io_bio; int skip_sum; bool write = (bio_op(dio_bio) == REQ_OP_WRITE); int ret = 0; skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - io_bio = btrfs_bio_clone(dio_bio, GFP_NOFS); - if (!io_bio) { + bio = btrfs_bio_clone(dio_bio, GFP_NOFS); + if (!bio) { ret = -ENOMEM; goto free_ordered; } @@ -8602,17 +8602,17 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct 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; - io_bio->bi_private = dip; - dip->orig_bio = io_bio; + bio->bi_private = dip; + dip->orig_bio = bio; dip->dio_bio = dio_bio; atomic_set(&dip->pending_bios, 0); - btrfs_bio = btrfs_io_bio(io_bio); - btrfs_bio->logical = file_offset; + io_bio = btrfs_io_bio(bio); + io_bio->logical = file_offset; if (write) { - io_bio->bi_end_io = btrfs_endio_direct_write; + bio->bi_end_io = btrfs_endio_direct_write; } else { - io_bio->bi_end_io = btrfs_endio_direct_read; + bio->bi_end_io = btrfs_endio_direct_read; dip->subio_endio = btrfs_subio_endio_read; } @@ -8635,8 +8635,8 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode, if (!ret) return; - if (btrfs_bio->end_io) - btrfs_bio->end_io(btrfs_bio, ret); + if (io_bio->end_io) + io_bio->end_io(io_bio, ret); free_ordered: /* @@ -8648,16 +8648,16 @@ free_ordered: * 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 (io_bio && dip) { - io_bio->bi_error = -EIO; - bio_endio(io_bio); + if (bio && dip) { + bio->bi_error = -EIO; + bio_endio(bio); /* - * The end io callbacks free our dip, do the final put on io_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; - io_bio = NULL; + bio = NULL; } else { if (write) __endio_write_update_ordered(inode, @@ -8675,8 +8675,8 @@ free_ordered: */ dio_end_io(dio_bio, ret); } - if (io_bio) - bio_put(io_bio); + if (bio) + bio_put(bio); kfree(dip); } -- cgit v1.2.3 From 3c91ee6964114b1a4f54ae6b397c889576d91cf4 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 18 May 2017 15:33:29 +0200 Subject: Btrfs: work around maybe-uninitialized warning A rewrite of btrfs_submit_direct_hook appears to have introduced a warning: fs/btrfs/inode.c: In function 'btrfs_submit_direct_hook': fs/btrfs/inode.c:8467:14: error: 'bio' may be used uninitialized in this function [-Werror=maybe-uninitialized] Where the 'bio' variable was previously initialized unconditionally, it is now set in the "while (submit_len > 0)" loop that would never execute if submit_len is zero. Assuming this cannot happen in practice, we can avoid the warning by simply replacing the while{} loop with a do{}while() loop so the compiler knows that it will always be entered at least once. Fixes changes introduced in "Btrfs: use bio_clone_bioset_partial to simplify DIO submit". Signed-off-by: Arnd Bergmann Signed-off-by: David Sterba --- fs/btrfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 53f2cf4b24b6..93f1b31199d2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8507,7 +8507,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, /* bio split */ ASSERT(map_length <= INT_MAX); atomic_inc(&dip->pending_bios); - while (submit_len > 0) { + do { clone_len = min_t(int, submit_len, map_length); /* @@ -8550,7 +8550,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, start_sector << 9, &map_length, NULL, 0); if (ret) goto out_err; - } + } while (submit_len > 0); submit: ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, -- cgit v1.2.3 From e477094f0d3ce35b30d230bda3f31fc060cfe93b Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Tue, 16 May 2017 10:57:14 -0700 Subject: Btrfs: hardcode GFP_NOFS for btrfs_bio_clone_partial We only pass GFP_NOFS to btrfs_bio_clone_partial, so lets hardcode it. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 5 ++--- fs/btrfs/extent_io.h | 3 +-- fs/btrfs/inode.c | 2 +- 3 files changed, 4 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index acb8c1d177e2..f2312b8aec4d 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2721,14 +2721,13 @@ struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) return bio; } -struct bio *btrfs_bio_clone_partial(struct bio *orig, gfp_t gfp_mask, - int offset, int size) +struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) { struct bio *bio; struct btrfs_io_bio *btrfs_bio; /* this will never fail when it's backed by a bioset */ - bio = bio_clone_fast(orig, gfp_mask, btrfs_bioset); + bio = bio_clone_fast(orig, GFP_NOFS, btrfs_bioset); ASSERT(bio); btrfs_bio = btrfs_io_bio(bio); diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 3df018549ce4..e3512c5d8770 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -467,8 +467,7 @@ btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, gfp_t gfp_flags); struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs); struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask); -struct bio *btrfs_bio_clone_partial(struct bio *orig, gfp_t gfp_mask, - int offset, int size); +struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size); struct btrfs_fs_info; struct btrfs_inode; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 93f1b31199d2..97b2bb9d3c89 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8514,7 +8514,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, * 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, GFP_NOFS, clone_offset, + bio = btrfs_bio_clone_partial(orig_bio, clone_offset, clone_len); bio->bi_private = dip; bio->bi_end_io = btrfs_end_dio_bio; -- cgit v1.2.3 From e03733da5aa68fa2ae3f8d1ab12a570c823a647f Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 12 May 2017 01:02:22 +0200 Subject: btrfs: fix bool type in btrfs_page_exists_in_range We use only a simple bool indicator, int is not a problem here. Signed-off-by: David Sterba --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 97b2bb9d3c89..683ee05798e5 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7486,7 +7486,7 @@ out: bool btrfs_page_exists_in_range(struct inode *inode, loff_t start, loff_t end) { struct radix_tree_root *root = &inode->i_mapping->page_tree; - int found = false; + bool found = false; void **pagep = NULL; struct page *page = NULL; unsigned long start_idx; -- cgit v1.2.3 From 1b86826d12dc4acf9576cad9c43da74025dc8074 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Wed, 17 May 2017 11:38:35 -0400 Subject: btrfs: cleanup root usage by btrfs_get_alloc_profile There are two places where we don't already know what kind of alloc profile we need before calling btrfs_get_alloc_profile, but we need access to a root everywhere we call it. This patch adds helpers for btrfs_{data,metadata,system}_alloc_profile() and relegates btrfs_system_alloc_profile to a static for use in those two cases. The next patch will eliminate one of those. Signed-off-by: Jeff Mahoney Reviewed-by: Liu Bo Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 4 +++- fs/btrfs/extent-tree.c | 28 +++++++++++++++++++++------- fs/btrfs/inode.c | 3 +-- fs/btrfs/super.c | 3 +-- fs/btrfs/volumes.c | 5 ++--- 5 files changed, 28 insertions(+), 15 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index c457cb177340..c13b4859df73 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2681,7 +2681,9 @@ void btrfs_get_block_group_trimming(struct btrfs_block_group_cache *cache); void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *cache); void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info); -u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data); +u64 btrfs_data_alloc_profile(struct btrfs_fs_info *fs_info); +u64 btrfs_metadata_alloc_profile(struct btrfs_fs_info *fs_info); +u64 btrfs_system_alloc_profile(struct btrfs_fs_info *fs_info); void btrfs_clear_space_info_full(struct btrfs_fs_info *info); enum btrfs_reserve_flush_enum { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 4c0d3980fe3f..47eb98ffdfc6 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4121,7 +4121,7 @@ static u64 get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags) return btrfs_reduce_alloc_profile(fs_info, flags); } -u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data) +static u64 get_alloc_profile_by_root(struct btrfs_root *root, int data) { struct btrfs_fs_info *fs_info = root->fs_info; u64 flags; @@ -4138,6 +4138,21 @@ u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data) return ret; } +u64 btrfs_data_alloc_profile(struct btrfs_fs_info *fs_info) +{ + return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_DATA); +} + +u64 btrfs_metadata_alloc_profile(struct btrfs_fs_info *fs_info) +{ + return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_METADATA); +} + +u64 btrfs_system_alloc_profile(struct btrfs_fs_info *fs_info) +{ + return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_SYSTEM); +} + static u64 btrfs_space_info_used(struct btrfs_space_info *s_info, bool may_use_included) { @@ -4187,7 +4202,7 @@ again: data_sinfo->force_alloc = CHUNK_ALLOC_FORCE; spin_unlock(&data_sinfo->lock); alloc: - alloc_target = btrfs_get_alloc_profile(root, 1); + alloc_target = btrfs_data_alloc_profile(fs_info); /* * It is ugly that we don't call nolock join * transaction for the free space inode case here. @@ -4463,9 +4478,8 @@ void check_system_chunk(struct btrfs_trans_handle *trans, } if (left < thresh) { - u64 flags; + u64 flags = btrfs_system_alloc_profile(fs_info); - flags = btrfs_get_alloc_profile(fs_info->chunk_root, 0); /* * Ignore failure to create system chunk. We might end up not * needing it, as we might not need to COW all nodes/leafs from @@ -4629,7 +4643,7 @@ static int can_overcommit(struct btrfs_root *root, if (space_info->flags & BTRFS_BLOCK_GROUP_DATA) return 0; - profile = btrfs_get_alloc_profile(root, 0); + profile = get_alloc_profile_by_root(root, 0); used = btrfs_space_info_used(space_info, false); /* @@ -4894,7 +4908,7 @@ static int flush_space(struct btrfs_fs_info *fs_info, break; } ret = do_chunk_alloc(trans, fs_info, - btrfs_get_alloc_profile(root, 0), + btrfs_metadata_alloc_profile(fs_info), CHUNK_ALLOC_NO_FORCE); btrfs_end_transaction(trans); if (ret > 0 || ret == -ENOSPC) @@ -7954,7 +7968,7 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, u64 flags; int ret; - flags = btrfs_get_alloc_profile(root, is_data); + flags = get_alloc_profile_by_root(root, is_data); again: WARN_ON(num_bytes < fs_info->sectorsize); ret = find_free_extent(fs_info, ram_bytes, num_bytes, empty_size, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 683ee05798e5..222d33d78092 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8473,7 +8473,6 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, { struct inode *inode = dip->inode; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_root *root = BTRFS_I(inode)->root; struct bio *bio; struct bio *orig_bio = dip->orig_bio; u64 start_sector = orig_bio->bi_iter.bi_sector; @@ -8499,7 +8498,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, } /* async crcs make it difficult to collect full stripe writes. */ - if (btrfs_get_alloc_profile(root, 1) & BTRFS_BLOCK_GROUP_RAID56_MASK) + if (btrfs_data_alloc_profile(fs_info) & BTRFS_BLOCK_GROUP_RAID56_MASK) async_submit = 0; else async_submit = 1; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 4f1cdd5058f1..3371213924bd 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1898,7 +1898,6 @@ static inline void btrfs_descending_sort_devices( static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info, u64 *free_bytes) { - struct btrfs_root *root = fs_info->tree_root; struct btrfs_device_info *devices_info; struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; struct btrfs_device *device; @@ -1932,7 +1931,7 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info, return -ENOMEM; /* calc min stripe number for data space allocation */ - type = btrfs_get_alloc_profile(root, 1); + type = btrfs_data_alloc_profile(fs_info); if (type & BTRFS_BLOCK_GROUP_RAID0) { min_stripes = 2; num_stripes = nr_devices; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e37f95976443..e28c113785bb 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -5019,20 +5019,19 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, static noinline int init_first_rw_device(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info) { - struct btrfs_root *extent_root = fs_info->extent_root; u64 chunk_offset; u64 sys_chunk_offset; u64 alloc_profile; int ret; chunk_offset = find_next_chunk(fs_info); - alloc_profile = btrfs_get_alloc_profile(extent_root, 0); + alloc_profile = btrfs_metadata_alloc_profile(fs_info); ret = __btrfs_alloc_chunk(trans, chunk_offset, alloc_profile); if (ret) return ret; sys_chunk_offset = find_next_chunk(fs_info); - alloc_profile = btrfs_get_alloc_profile(fs_info->chunk_root, 0); + alloc_profile = btrfs_system_alloc_profile(fs_info); ret = __btrfs_alloc_chunk(trans, sys_chunk_offset, alloc_profile); return ret; } -- cgit v1.2.3 From 04a87e3472828f769a93655d7c64a27573bdbc2c Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Fri, 12 May 2017 15:07:43 -0700 Subject: Btrfs: add statx support Return enhanced file attributes from the btrfs, including: (1). inode creation time as stx_btime, and (2). Certain BTRFS_INODE_xxx flags are mapped to stx_attributes flags. Example output: [root@localhost ~]# cat t.sh touch t chattr +aic t ~/linux/samples/statx/test-statx t chattr -aic t touch t echo "========================================" ~/linux/samples/statx/test-statx t /bin/rm t [root@localhost ~]# ./t.sh statx(t) = 0 results=fff Size: 0 Blocks: 0 IO Block: 4096 regular file Device: 00:1c Inode: 63962 Links: 1 Access: (0644/-rw-r--r--) Uid: 0 Gid: 0 Access: 2017-05-11 16:03:13.999856591-0700 Modify: 2017-05-11 16:03:13.999856591-0700 Change: 2017-05-11 16:03:14.000856663-0700 Birth: 2017-05-11 16:03:13.999856591-0700 Attributes: 0000000000000034 (........ ........ ........ ........ ........ ........ ........ .-ai.c..) ======================================== statx(t) = 0 results=fff Size: 0 Blocks: 0 IO Block: 4096 regular file Device: 00:1c Inode: 63962 Links: 1 Access: (0644/-rw-r--r--) Uid: 0 Gid: 0 Access: 2017-05-11 16:03:14.006857097-0700 Modify: 2017-05-11 16:03:14.006857097-0700 Change: 2017-05-11 16:03:14.006857097-0700 Birth: 2017-05-11 16:03:13.999856591-0700 Attributes: 0000000000000000 (........ ........ ........ ........ ........ ........ ........ .---.-..) [root@localhost ~]# Reviewed-by: Omar Sandoval Signed-off-by: Yonghong Song Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 222d33d78092..6b8d34136044 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9563,6 +9563,24 @@ static int btrfs_getattr(const struct path *path, struct kstat *stat, u64 delalloc_bytes; struct inode *inode = d_inode(path->dentry); u32 blocksize = inode->i_sb->s_blocksize; + u32 bi_flags = BTRFS_I(inode)->flags; + + stat->result_mask |= STATX_BTIME; + stat->btime.tv_sec = BTRFS_I(inode)->i_otime.tv_sec; + stat->btime.tv_nsec = BTRFS_I(inode)->i_otime.tv_nsec; + if (bi_flags & BTRFS_INODE_APPEND) + stat->attributes |= STATX_ATTR_APPEND; + if (bi_flags & BTRFS_INODE_COMPRESS) + stat->attributes |= STATX_ATTR_COMPRESSED; + if (bi_flags & BTRFS_INODE_IMMUTABLE) + stat->attributes |= STATX_ATTR_IMMUTABLE; + if (bi_flags & BTRFS_INODE_NODUMP) + stat->attributes |= STATX_ATTR_NODUMP; + + stat->attributes_mask |= (STATX_ATTR_APPEND | + STATX_ATTR_COMPRESSED | + STATX_ATTR_IMMUTABLE | + STATX_ATTR_NODUMP); generic_fillattr(inode, stat); stat->dev = BTRFS_I(inode)->root->anon_dev; -- cgit v1.2.3 From 4b5faeac4688174fd523f2a22b7d70d5a96842fb Mon Sep 17 00:00:00 2001 From: David Sterba Date: Tue, 28 Mar 2017 12:06:05 +0200 Subject: btrfs: use generic slab for for btrfs_transaction Observing the number of slab objects of btrfs_transaction, there's just one active on an almost quiescent filesystem, and the number of objects goes to about ten when sync is in progress. Then the nubmer goes down to 1. This matches the expectations of the transaction lifetime. For such use the separate slab cache is not justified, as we do not reuse objects frequently. For the shortlived transaction, the generic slab (size 512) should be ok. We can optimistically expect that the 512 slabs are not all used (fragmentation) and there are free slots to take when we do the allocation, compared to potentially allocating a whole new page for the separate slab. We'll lose the stats about the object use, which could be added later if we really need them. Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 1 - fs/btrfs/disk-io.c | 5 ----- fs/btrfs/inode.c | 8 -------- fs/btrfs/transaction.c | 8 ++++---- 4 files changed, 4 insertions(+), 18 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 388f1128544e..f0f5f28784b6 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -48,7 +48,6 @@ struct btrfs_trans_handle; struct btrfs_transaction; struct btrfs_pending_snapshot; extern struct kmem_cache *btrfs_trans_handle_cachep; -extern struct kmem_cache *btrfs_transaction_cachep; extern struct kmem_cache *btrfs_bit_radix_cachep; extern struct kmem_cache *btrfs_path_cachep; extern struct kmem_cache *btrfs_free_space_cachep; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index cdc28ad7f1ed..cb79bce3a972 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4624,11 +4624,6 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans, cur_trans->state =TRANS_STATE_COMPLETED; wake_up(&cur_trans->commit_wait); - - /* - memset(cur_trans, 0, sizeof(*cur_trans)); - kmem_cache_free(btrfs_transaction_cachep, cur_trans); - */ } static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6b8d34136044..0a1ec5cd3b8f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -86,7 +86,6 @@ static const struct extent_io_ops btrfs_extent_io_ops; static struct kmem_cache *btrfs_inode_cachep; struct kmem_cache *btrfs_trans_handle_cachep; -struct kmem_cache *btrfs_transaction_cachep; struct kmem_cache *btrfs_path_cachep; struct kmem_cache *btrfs_free_space_cachep; @@ -9513,7 +9512,6 @@ void btrfs_destroy_cachep(void) rcu_barrier(); kmem_cache_destroy(btrfs_inode_cachep); kmem_cache_destroy(btrfs_trans_handle_cachep); - kmem_cache_destroy(btrfs_transaction_cachep); kmem_cache_destroy(btrfs_path_cachep); kmem_cache_destroy(btrfs_free_space_cachep); } @@ -9533,12 +9531,6 @@ int btrfs_init_cachep(void) if (!btrfs_trans_handle_cachep) goto fail; - btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction", - sizeof(struct btrfs_transaction), 0, - SLAB_TEMPORARY | SLAB_MEM_SPREAD, NULL); - if (!btrfs_transaction_cachep) - goto fail; - btrfs_path_cachep = kmem_cache_create("btrfs_path", sizeof(struct btrfs_path), 0, SLAB_MEM_SPREAD, NULL); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index ca0009ff47f1..ab030fb22530 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -93,7 +93,7 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction) btrfs_put_block_group_trimming(cache); btrfs_put_block_group(cache); } - kmem_cache_free(btrfs_transaction_cachep, transaction); + kfree(transaction); } } @@ -228,7 +228,7 @@ loop: */ BUG_ON(type == TRANS_JOIN_NOLOCK); - cur_trans = kmem_cache_alloc(btrfs_transaction_cachep, GFP_NOFS); + cur_trans = kmalloc(sizeof(*cur_trans), GFP_NOFS); if (!cur_trans) return -ENOMEM; @@ -238,11 +238,11 @@ loop: * someone started a transaction after we unlocked. Make sure * to redo the checks above */ - kmem_cache_free(btrfs_transaction_cachep, cur_trans); + kfree(cur_trans); goto loop; } else if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) { spin_unlock(&fs_info->trans_lock); - kmem_cache_free(btrfs_transaction_cachep, cur_trans); + kfree(cur_trans); return -EROFS; } -- cgit v1.2.3 From 054ec2f626ed5bc1633addfabb1d8de1b56f949d Mon Sep 17 00:00:00 2001 From: Guoqing Jiang Date: Fri, 2 Jun 2017 16:08:50 +0800 Subject: btrfs: simplify code with bio_io_error bio_io_error was introduced in the commit 4246a0b63bd8f56a1469b ("block: add a bi_error field to struct bio"), so use it to simplify code. Signed-off-by: Guoqing Jiang Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 0a1ec5cd3b8f..4e33c3fded5c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8647,8 +8647,7 @@ free_ordered: * callbacks - they require an allocated dip and a clone of dio_bio. */ if (bio && dip) { - bio->bi_error = -EIO; - bio_endio(bio); + 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 -- cgit v1.2.3 From 3aa8e074ab2ec507b7380dd5d460fea83d92b66b Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 2 Jun 2017 17:38:30 +0200 Subject: btrfs: btrfs_bio_clone never fails, skip error handling Update direct callers of btrfs_bio_clone that do error handling, that we can now remove. Reviewed-by: Anand Jain Signed-off-by: David Sterba --- fs/btrfs/inode.c | 4 ---- fs/btrfs/volumes.c | 5 ++--- 2 files changed, 2 insertions(+), 7 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4e33c3fded5c..1fde50a430a9 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8584,10 +8584,6 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode, skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; bio = btrfs_bio_clone(dio_bio, GFP_NOFS); - if (!bio) { - ret = -ENOMEM; - goto free_ordered; - } dip = kzalloc(sizeof(*dip), GFP_NOFS); if (!dip) { diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e28c113785bb..ea9912157a56 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6255,10 +6255,9 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, continue; } - if (dev_nr < total_devs - 1) { + if (dev_nr < total_devs - 1) bio = btrfs_bio_clone(first_bio, GFP_NOFS); - BUG_ON(!bio); /* -ENOMEM */ - } else + else bio = first_bio; submit_stripe_bio(bbio, bio, bbio->stripes[dev_nr].physical, -- cgit v1.2.3 From 8b6c1d56f2f5094b14b22a226b798ca3d186c0e9 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 2 Jun 2017 17:48:13 +0200 Subject: btrfs: sink gfp parameter to btrfs_bio_clone All callers pass GFP_NOFS. Reviewed-by: Anand Jain Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 4 ++-- fs/btrfs/extent_io.h | 2 +- fs/btrfs/inode.c | 2 +- fs/btrfs/volumes.c | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index bfcbe8f2818b..d60221103ff4 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2675,13 +2675,13 @@ btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, return bio; } -struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask) +struct bio *btrfs_bio_clone(struct bio *bio) { struct btrfs_io_bio *btrfs_bio; struct bio *new; /* Bio allocation backed by a bioset does not fail */ - new = bio_clone_fast(bio, gfp_mask, btrfs_bioset); + new = bio_clone_fast(bio, GFP_NOFS, btrfs_bioset); btrfs_bio = btrfs_io_bio(new); btrfs_bio->csum = NULL; btrfs_bio->csum_allocated = NULL; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index e3512c5d8770..4fe643a5aeaf 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -466,7 +466,7 @@ struct bio * btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, gfp_t gfp_flags); struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs); -struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask); +struct bio *btrfs_bio_clone(struct bio *bio); struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size); struct btrfs_fs_info; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1fde50a430a9..96bbc836d071 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8583,7 +8583,7 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode, skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - bio = btrfs_bio_clone(dio_bio, GFP_NOFS); + bio = btrfs_bio_clone(dio_bio); dip = kzalloc(sizeof(*dip), GFP_NOFS); if (!dip) { diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index ea9912157a56..c10d75fb2202 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6256,7 +6256,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, } if (dev_nr < total_devs - 1) - bio = btrfs_bio_clone(first_bio, GFP_NOFS); + bio = btrfs_bio_clone(first_bio); else bio = first_bio; -- cgit v1.2.3 From e3d37faba2eb19a1d459917bbf54ac1c65711510 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Wed, 17 May 2017 15:42:00 -0600 Subject: Btrfs: tolerate errors if we have retried successfully With raid1 profile, dio read isn't tolerating IO errors if read length is less than the stripe length (64K). Our bio didn't get split in btrfs_submit_direct_hook() if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED) is true and that happens when the read length is less than 64k. In this case, if the underlying device returns error somehow, bio->bi_error has recorded that error. If we could recover the correct data from another copy in profile raid1/10/5/6, with btrfs_subio_endio_read() returning 0, bio would have the correct data in its vector, but bio->bi_error is not updated accordingly so that the following dio_end_io(dio_bio, bio->bi_error) makes directIO think this read has failed. This fixes the problem by setting bio's error to 0 if a good copy has been found. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 96bbc836d071..147bafdb8628 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8261,8 +8261,11 @@ static void btrfs_endio_direct_read(struct bio *bio) struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); int err = bio->bi_error; - if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED) + if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED) { err = btrfs_subio_endio_read(inode, io_bio, err); + if (!err) + bio->bi_error = 0; + } unlock_extent(&BTRFS_I(inode)->io_tree, dip->logical_offset, dip->logical_offset + dip->bytes - 1); -- cgit v1.2.3 From ef7cdac101602844a4179f7d2d9e26bd20a0edf9 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Thu, 13 Apr 2017 18:11:48 -0700 Subject: Btrfs: skip checksum verification if IO error occurs Currently dio read also goes to verify checksum if -EIO has been returned, although it usually fails on checksum, it's not necessary at all, we could directly check if there is another copy to read. And with this, the behavior of dio read is now consistent with that of buffered read. Signed-off-by: Liu Bo Reviewed-by: David Sterba [ use bool for uptodate ] Signed-off-by: David Sterba --- fs/btrfs/inode.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 147bafdb8628..b74a6d23a0fe 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8180,6 +8180,7 @@ static int __btrfs_subio_endio_read(struct inode *inode, int nr_sectors; unsigned int pgoff; int csum_pos; + bool uptodate = (err == 0); int ret; fs_info = BTRFS_I(inode)->root->fs_info; @@ -8195,12 +8196,13 @@ static int __btrfs_subio_endio_read(struct inode *inode, pgoff = bvec.bv_offset; next_block: - 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; + 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; -- cgit v1.2.3 From 170607ebd9c891d6765445434460065b2e73ca51 Mon Sep 17 00:00:00 2001 From: Timofey Titovets Date: Tue, 6 Jun 2017 14:41:15 +0300 Subject: Btrfs: compression must free at least one sector size We already skip storing data where compression does not make the result at least one byte less. Let's make the logic better and check that compression frees at least one sector size of bytes, otherwise it's not that useful. Signed-off-by: Timofey Titovets Reviewed-by: David Sterba [ changelog updated ] Signed-off-by: David Sterba --- fs/btrfs/inode.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index b74a6d23a0fe..a21a984d84d9 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -607,12 +607,11 @@ cont: /* * one last check to make sure the compression is really a - * win, compare the page count read with the blocks on disk + * 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); - if (total_compressed >= total_in) { - will_compress = 0; - } else { + if (total_compressed + blocksize <= total_in) { num_bytes = total_in; *num_added += 1; -- cgit v1.2.3 From 79b4f4c605b76194448315865ebcf6fcb0844fc5 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 15 Jun 2017 19:09:51 +0200 Subject: btrfs: cleanup duplicate return value in insert_inline_extent The pattern when err is used for function exit and ret is used for return values of callees is not used here. Reviewed-by: Nikolay Borisov Signed-off-by: David Sterba --- fs/btrfs/inode.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a21a984d84d9..e837713d0d05 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -177,7 +177,6 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, char *kaddr; unsigned long ptr; struct btrfs_file_extent_item *ei; - int err = 0; int ret; size_t cur_size = size; unsigned long offset; @@ -199,10 +198,8 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, path->leave_spinning = 1; ret = btrfs_insert_empty_item(trans, root, path, &key, datasize); - if (ret) { - err = ret; + if (ret) goto fail; - } } leaf = path->nodes[0]; ei = btrfs_item_ptr(leaf, path->slots[0], @@ -257,9 +254,8 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, BTRFS_I(inode)->disk_i_size = inode->i_size; ret = btrfs_update_inode(trans, root, inode); - return ret; fail: - return err; + return ret; } -- cgit v1.2.3 From e79a33270d05f711e985b9524a392fd45ad3e93f Mon Sep 17 00:00:00 2001 From: Su Yue Date: Tue, 6 Jun 2017 17:57:01 +0800 Subject: btrfs: Check name_len with boundary in verify dir_item Originally, verify_dir_item verifies name_len of dir_item with fixed values but not item boundary. If corrupted name_len was not bigger than the fixed value, for example 255, the function will think the dir_item is fine. And then reading beyond boundary will cause crash. Example: 1. Corrupt one dir_item name_len to be 255. 2. Run 'ls -lar /mnt/test/ > /dev/null' dmesg: [ 48.451449] BTRFS info (device vdb1): disk space caching is enabled [ 48.451453] BTRFS info (device vdb1): has skinny extents [ 48.489420] general protection fault: 0000 [#1] SMP [ 48.489571] Modules linked in: ext4 jbd2 mbcache btrfs xor raid6_pq [ 48.489716] CPU: 1 PID: 2710 Comm: ls Not tainted 4.10.0-rc1 #5 [ 48.489853] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-20170228_101828-anatol 04/01/2014 [ 48.490008] task: ffff880035df1bc0 task.stack: ffffc90004800000 [ 48.490008] RIP: 0010:read_extent_buffer+0xd2/0x190 [btrfs] [ 48.490008] RSP: 0018:ffffc90004803d98 EFLAGS: 00010202 [ 48.490008] RAX: 000000000000001b RBX: 000000000000001b RCX: 0000000000000000 [ 48.490008] RDX: ffff880079dbf36c RSI: 0005080000000000 RDI: ffff880079dbf368 [ 48.490008] RBP: ffffc90004803dc8 R08: ffff880078e8cc48 R09: ffff880000000000 [ 48.490008] R10: 0000160000000000 R11: 0000000000001000 R12: ffff880079dbf288 [ 48.490008] R13: ffff880078e8ca88 R14: 0000000000000003 R15: ffffc90004803e20 [ 48.490008] FS: 00007fef50c60800(0000) GS:ffff88007d400000(0000) knlGS:0000000000000000 [ 48.490008] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 48.490008] CR2: 000055f335ac2ff8 CR3: 000000007356d000 CR4: 00000000001406e0 [ 48.490008] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 48.490008] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 48.490008] Call Trace: [ 48.490008] btrfs_real_readdir+0x3b7/0x4a0 [btrfs] [ 48.490008] iterate_dir+0x181/0x1b0 [ 48.490008] SyS_getdents+0xa7/0x150 [ 48.490008] ? fillonedir+0x150/0x150 [ 48.490008] entry_SYSCALL_64_fastpath+0x18/0xad [ 48.490008] RIP: 0033:0x7fef5032546b [ 48.490008] RSP: 002b:00007ffeafcdb830 EFLAGS: 00000206 ORIG_RAX: 000000000000004e [ 48.490008] RAX: ffffffffffffffda RBX: 00007fef5061db38 RCX: 00007fef5032546b [ 48.490008] RDX: 0000000000008000 RSI: 000055f335abaff0 RDI: 0000000000000003 [ 48.490008] RBP: 00007fef5061dae0 R08: 00007fef5061db48 R09: 0000000000000000 [ 48.490008] R10: 000055f335abafc0 R11: 0000000000000206 R12: 00007fef5061db38 [ 48.490008] R13: 0000000000008040 R14: 00007fef5061db38 R15: 000000000000270e [ 48.490008] RIP: read_extent_buffer+0xd2/0x190 [btrfs] RSP: ffffc90004803d98 [ 48.499455] ---[ end trace 321920d8e8339505 ]--- Fix it by adding a parameter @slot and check name_len with item boundary by calling btrfs_is_name_len_valid. Signed-off-by: Su Yue rev Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 2 +- fs/btrfs/dir-item.c | 10 +++++++++- fs/btrfs/inode.c | 2 +- fs/btrfs/tree-log.c | 4 ++-- fs/btrfs/xattr.c | 2 +- 5 files changed, 14 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index d182ce792173..5e33e1d6d5c9 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3036,7 +3036,7 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans, const char *name, u16 name_len, int mod); int verify_dir_item(struct btrfs_fs_info *fs_info, - struct extent_buffer *leaf, + struct extent_buffer *leaf, int slot, struct btrfs_dir_item *dir_item); struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info, struct btrfs_path *path, diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index 5b6c6fb7c800..d9c4a3dd071e 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -395,7 +395,7 @@ struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info, leaf = path->nodes[0]; dir_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dir_item); - if (verify_dir_item(fs_info, leaf, dir_item)) + if (verify_dir_item(fs_info, leaf, path->slots[0], dir_item)) return NULL; total_len = btrfs_item_size_nr(leaf, path->slots[0]); @@ -453,9 +453,11 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans, int verify_dir_item(struct btrfs_fs_info *fs_info, struct extent_buffer *leaf, + int slot, struct btrfs_dir_item *dir_item) { u16 namelen = BTRFS_NAME_LEN; + int ret; u8 type = btrfs_dir_type(leaf, dir_item); if (type >= BTRFS_FT_MAX) { @@ -472,6 +474,12 @@ int verify_dir_item(struct btrfs_fs_info *fs_info, return 1; } + namelen = btrfs_dir_name_len(leaf, dir_item); + ret = btrfs_is_name_len_valid(leaf, slot, + (unsigned long)(dir_item + 1), namelen); + if (!ret) + return 1; + /* BTRFS_MAX_XATTR_SIZE is the same for all dir items */ if ((btrfs_dir_data_len(leaf, dir_item) + btrfs_dir_name_len(leaf, dir_item)) > diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e837713d0d05..86cac431864b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5934,7 +5934,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) ctx->pos = found_key.offset; di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item); - if (verify_dir_item(fs_info, leaf, di)) + if (verify_dir_item(fs_info, leaf, slot, di)) goto next; name_len = btrfs_dir_name_len(leaf, di); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index ccfe9fe7754a..1930f28edcdd 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1841,7 +1841,7 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans, ptr_end = ptr + item_size; while (ptr < ptr_end) { di = (struct btrfs_dir_item *)ptr; - if (verify_dir_item(fs_info, eb, di)) + if (verify_dir_item(fs_info, eb, slot, di)) return -EIO; name_len = btrfs_dir_name_len(eb, di); ret = replay_one_name(trans, root, path, eb, di, key); @@ -2017,7 +2017,7 @@ again: ptr_end = ptr + item_size; while (ptr < ptr_end) { di = (struct btrfs_dir_item *)ptr; - if (verify_dir_item(fs_info, eb, di)) { + if (verify_dir_item(fs_info, eb, slot, di)) { ret = -EIO; goto out; } diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index b3cbf80c5acf..2c7e53f9ff1b 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -336,7 +336,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) u32 this_len = sizeof(*di) + name_len + data_len; unsigned long name_ptr = (unsigned long)(di + 1); - if (verify_dir_item(fs_info, leaf, di)) { + if (verify_dir_item(fs_info, leaf, slot, di)) { ret = -EIO; goto err; } -- cgit v1.2.3 From a12b877b557dde885bdb6978cec86bf761e63c9a Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 27 Feb 2017 15:10:37 +0800 Subject: btrfs: qgroup: Fix qgroup reserved space underflow caused by buffered write and quotas being enabled [BUG] Under the following case, we can underflow qgroup reserved space. Task A | Task B --------------------------------------------------------------- Quota disabled | Buffered write | |- btrfs_check_data_free_space() | | *NO* qgroup space is reserved | | since quota is *DISABLED* | |- All pages are copied to page | cache | | Enable quota | Quota scan finished | | Sync_fs | |- run_delalloc_range | |- Write pages | |- btrfs_finish_ordered_io | |- insert_reserved_file_extent | |- btrfs_qgroup_release_data() | Since no qgroup space is reserved in Task A, we underflow qgroup reserved space This can be detected by fstest btrfs/104. [CAUSE] In insert_reserved_file_extent() we tell qgroup to release the @ram_bytes size of qgroup reserved_space in all cases. And btrfs_qgroup_release_data() will check if quotas are enabled. However in the above case, the buffered write happens before quota is enabled, so we don't have the reserved space for that range. [FIX] In insert_reserved_file_extent(), we tell qgroup to release the acctual byte number it released. In the above case, since we don't have the reserved space, we tell qgroups to release 0 byte, so the problem can be fixed. And thanks to the @reserved parameter introduced by the qgroup rework, and previous patch to return released bytes, the fix can be as small as 10 lines. Signed-off-by: Qu Wenruo [ changelog updates ] Signed-off-by: David Sterba --- fs/btrfs/inode.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 86cac431864b..b66ea03a3a1c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2145,6 +2145,7 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_key ins; + u64 qg_released; int extent_inserted = 0; int ret; @@ -2200,13 +2201,17 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, ins.objectid = disk_bytenr; ins.offset = disk_num_bytes; ins.type = BTRFS_EXTENT_ITEM_KEY; - ret = btrfs_alloc_reserved_file_extent(trans, root->root_key.objectid, - btrfs_ino(BTRFS_I(inode)), file_pos, ram_bytes, &ins); + /* * Release the reserved range from inode dirty range map, as it is * already moved into delayed_ref_head */ - btrfs_qgroup_release_data(inode, file_pos, ram_bytes); + ret = btrfs_qgroup_release_data(inode, file_pos, ram_bytes); + if (ret < 0) + goto out; + qg_released = ret; + ret = btrfs_alloc_reserved_file_extent(trans, root->root_key.objectid, + btrfs_ino(BTRFS_I(inode)), file_pos, qg_released, &ins); out: btrfs_free_path(path); -- cgit v1.2.3 From 364ecf3651e0862152c8b340d7cb3021dc0122c7 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 27 Feb 2017 15:10:38 +0800 Subject: btrfs: qgroup: Introduce extent changeset for qgroup reserve functions Introduce a new parameter, struct extent_changeset for btrfs_qgroup_reserved_data() and its callers. Such extent_changeset was used in btrfs_qgroup_reserve_data() to record which range it reserved in current reserve, so it can free it in error paths. The reason we need to export it to callers is, at buffered write error path, without knowing what exactly which range we reserved in current allocation, we can free space which is not reserved by us. This will lead to qgroup reserved space underflow. Reviewed-by: Chandan Rajendra Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 6 ++++-- fs/btrfs/extent-tree.c | 23 +++++++++++++---------- fs/btrfs/extent_io.h | 34 +++++++++++++++++++++++++++++++++ fs/btrfs/file.c | 12 +++++++++--- fs/btrfs/inode-map.c | 4 +++- fs/btrfs/inode.c | 18 ++++++++++++++---- fs/btrfs/ioctl.c | 5 ++++- fs/btrfs/qgroup.c | 51 ++++++++++++++++++++++++++++++++------------------ fs/btrfs/qgroup.h | 3 ++- fs/btrfs/relocation.c | 4 +++- 10 files changed, 119 insertions(+), 41 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 5e33e1d6d5c9..1ee4489dc398 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2708,8 +2708,9 @@ enum btrfs_flush_state { COMMIT_TRANS = 6, }; -int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len); int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes); +int btrfs_check_data_free_space(struct inode *inode, + struct extent_changeset **reserved, u64 start, u64 len); void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len); void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start, u64 len); @@ -2727,7 +2728,8 @@ void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *rsv); int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes); void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes); -int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len); +int btrfs_delalloc_reserve_space(struct inode *inode, + struct extent_changeset **reserved, u64 start, u64 len); void btrfs_delalloc_release_space(struct inode *inode, u64 start, u64 len); void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type); struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info, diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index b0b02c6c71aa..70f85cfdbd46 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3402,6 +3402,7 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group, struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_root *root = fs_info->tree_root; struct inode *inode = NULL; + struct extent_changeset *data_reserved = NULL; u64 alloc_hint = 0; int dcs = BTRFS_DC_ERROR; u64 num_pages = 0; @@ -3521,7 +3522,7 @@ again: num_pages *= 16; num_pages *= PAGE_SIZE; - ret = btrfs_check_data_free_space(inode, 0, num_pages); + ret = btrfs_check_data_free_space(inode, &data_reserved, 0, num_pages); if (ret) goto out_put; @@ -3552,6 +3553,7 @@ out: block_group->disk_cache_state = dcs; spin_unlock(&block_group->lock); + extent_changeset_free(data_reserved); return ret; } @@ -4326,12 +4328,8 @@ commit_trans: return ret; } -/* - * New check_data_free_space() with ability for precious data reservation - * Will replace old btrfs_check_data_free_space(), but for patch split, - * add a new function first and then replace it. - */ -int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len) +int btrfs_check_data_free_space(struct inode *inode, + struct extent_changeset **reserved, u64 start, u64 len) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int ret; @@ -4346,9 +4344,11 @@ int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len) return ret; /* Use new btrfs_qgroup_reserve_data to reserve precious data space. */ - ret = btrfs_qgroup_reserve_data(inode, start, len); + ret = btrfs_qgroup_reserve_data(inode, reserved, start, len); if (ret < 0) btrfs_free_reserved_data_space_noquota(inode, start, len); + else + ret = 0; return ret; } @@ -6175,6 +6175,8 @@ void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes) * @inode: inode we're writing to * @start: start range we are writing to * @len: how long the range we are writing to + * @reserved: mandatory parameter, record actually reserved qgroup ranges of + * current reservation. * * This will do the following things * @@ -6192,11 +6194,12 @@ void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes) * Return 0 for success * Return <0 for error(-ENOSPC or -EQUOT) */ -int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len) +int btrfs_delalloc_reserve_space(struct inode *inode, + struct extent_changeset **reserved, u64 start, u64 len) { int ret; - ret = btrfs_check_data_free_space(inode, start, len); + ret = btrfs_check_data_free_space(inode, reserved, start, len); if (ret < 0) return ret; ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len); diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index ce670d213913..aeafdb35d90b 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -215,6 +215,40 @@ struct extent_changeset { struct ulist range_changed; }; +static inline void extent_changeset_init(struct extent_changeset *changeset) +{ + changeset->bytes_changed = 0; + ulist_init(&changeset->range_changed); +} + +static inline struct extent_changeset *extent_changeset_alloc(void) +{ + struct extent_changeset *ret; + + ret = kmalloc(sizeof(*ret), GFP_KERNEL); + if (!ret) + return NULL; + + extent_changeset_init(ret); + return ret; +} + +static inline void extent_changeset_release(struct extent_changeset *changeset) +{ + if (!changeset) + return; + changeset->bytes_changed = 0; + ulist_release(&changeset->range_changed); +} + +static inline void extent_changeset_free(struct extent_changeset *changeset) +{ + if (!changeset) + return; + extent_changeset_release(changeset); + kfree(changeset); +} + static inline void extent_set_compress_type(unsigned long *bio_flags, int compress_type) { diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 5da85b080368..1b5cce51728b 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1581,6 +1581,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, struct btrfs_root *root = BTRFS_I(inode)->root; struct page **pages = NULL; struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; u64 release_bytes = 0; u64 lockstart; u64 lockend; @@ -1628,7 +1629,9 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, reserve_bytes = round_up(write_bytes + sector_offset, fs_info->sectorsize); - ret = btrfs_check_data_free_space(inode, pos, write_bytes); + extent_changeset_release(data_reserved); + ret = btrfs_check_data_free_space(inode, &data_reserved, pos, + write_bytes); if (ret < 0) { if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)) && @@ -1802,6 +1805,7 @@ again: } } + extent_changeset_free(data_reserved); return num_written ? num_written : ret; } @@ -2772,6 +2776,7 @@ static long btrfs_fallocate(struct file *file, int mode, { struct inode *inode = file_inode(file); struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; struct falloc_range *range; struct falloc_range *tmp; struct list_head reserve_list; @@ -2901,8 +2906,8 @@ static long btrfs_fallocate(struct file *file, int mode, free_extent_map(em); break; } - ret = btrfs_qgroup_reserve_data(inode, cur_offset, - last_byte - cur_offset); + ret = btrfs_qgroup_reserve_data(inode, &data_reserved, + cur_offset, last_byte - cur_offset); if (ret < 0) { free_extent_map(em); break; @@ -2974,6 +2979,7 @@ out: if (ret != 0) btrfs_free_reserved_data_space(inode, alloc_start, alloc_end - cur_offset); + extent_changeset_free(data_reserved); return ret; } diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index 5c6c20ec64d8..d02019747d00 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -400,6 +400,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root, struct btrfs_path *path; struct inode *inode; struct btrfs_block_rsv *rsv; + struct extent_changeset *data_reserved = NULL; u64 num_bytes; u64 alloc_hint = 0; int ret; @@ -492,7 +493,7 @@ again: /* Just to make sure we have enough space */ prealloc += 8 * PAGE_SIZE; - ret = btrfs_delalloc_reserve_space(inode, 0, prealloc); + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, 0, prealloc); if (ret) goto out_put; @@ -516,6 +517,7 @@ out: trans->bytes_reserved = num_bytes; btrfs_free_path(path); + extent_changeset_free(data_reserved); return ret; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index b66ea03a3a1c..d9cf6df40d5e 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2037,6 +2037,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work) struct btrfs_writepage_fixup *fixup; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; struct page *page; struct inode *inode; u64 page_start; @@ -2074,7 +2075,7 @@ again: goto again; } - ret = btrfs_delalloc_reserve_space(inode, page_start, + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start, PAGE_SIZE); if (ret) { mapping_set_error(page->mapping, ret); @@ -2094,6 +2095,7 @@ out_page: unlock_page(page); put_page(page); kfree(fixup); + extent_changeset_free(data_reserved); } /* @@ -4769,6 +4771,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; char *kaddr; u32 blocksize = fs_info->sectorsize; pgoff_t index = from >> PAGE_SHIFT; @@ -4783,7 +4786,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, (!len || ((len & (blocksize - 1)) == 0))) goto out; - ret = btrfs_delalloc_reserve_space(inode, + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, round_down(from, blocksize), blocksize); if (ret) goto out; @@ -4868,6 +4871,7 @@ out_unlock: unlock_page(page); put_page(page); out: + extent_changeset_free(data_reserved); return ret; } @@ -8718,6 +8722,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) 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; @@ -8754,7 +8759,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) inode_unlock(inode); relock = true; } - ret = btrfs_delalloc_reserve_space(inode, offset, count); + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, + offset, count); if (ret) goto out; dio_data.outstanding_extents = count_max_extents(count); @@ -8811,6 +8817,7 @@ out: if (relock) inode_lock(inode); + extent_changeset_free(data_reserved); return ret; } @@ -9043,6 +9050,7 @@ int btrfs_page_mkwrite(struct vm_fault *vmf) struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; 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; @@ -9068,7 +9076,7 @@ int btrfs_page_mkwrite(struct vm_fault *vmf) * end up waiting indefinitely to get a lock on the page currently * being processed by btrfs_page_mkwrite() function. */ - ret = btrfs_delalloc_reserve_space(inode, page_start, + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start, reserved_space); if (!ret) { ret = file_update_time(vmf->vma->vm_file); @@ -9174,6 +9182,7 @@ again: out_unlock: if (!ret) { sb_end_pagefault(inode->i_sb); + extent_changeset_free(data_reserved); return VM_FAULT_LOCKED; } unlock_page(page); @@ -9181,6 +9190,7 @@ out: btrfs_delalloc_release_space(inode, page_start, reserved_space); out_noreserve: sb_end_pagefault(inode->i_sb); + extent_changeset_free(data_reserved); return ret; } diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index e4116f9248c2..ccee5417d3f6 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1127,6 +1127,7 @@ static int cluster_pages_for_defrag(struct inode *inode, struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; struct extent_io_tree *tree; + struct extent_changeset *data_reserved = NULL; gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); file_end = (isize - 1) >> PAGE_SHIFT; @@ -1135,7 +1136,7 @@ static int cluster_pages_for_defrag(struct inode *inode, page_cnt = min_t(u64, (u64)num_pages, (u64)file_end - start_index + 1); - ret = btrfs_delalloc_reserve_space(inode, + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, start_index << PAGE_SHIFT, page_cnt << PAGE_SHIFT); if (ret) @@ -1247,6 +1248,7 @@ again: unlock_page(pages[i]); put_page(pages[i]); } + extent_changeset_free(data_reserved); return i_done; out: for (i = 0; i < i_done; i++) { @@ -1256,6 +1258,7 @@ out: btrfs_delalloc_release_space(inode, start_index << PAGE_SHIFT, page_cnt << PAGE_SHIFT); + extent_changeset_free(data_reserved); return ret; } diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 475d53c492c8..002088937021 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -2835,43 +2835,60 @@ btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info) * Return <0 for error (including -EQUOT) * * NOTE: this function may sleep for memory allocation. + * if btrfs_qgroup_reserve_data() is called multiple times with + * same @reserved, caller must ensure when error happens it's OK + * to free *ALL* reserved space. */ -int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len) +int btrfs_qgroup_reserve_data(struct inode *inode, + struct extent_changeset **reserved_ret, u64 start, + u64 len) { struct btrfs_root *root = BTRFS_I(inode)->root; - struct extent_changeset changeset; struct ulist_node *unode; struct ulist_iterator uiter; + struct extent_changeset *reserved; + u64 orig_reserved; + u64 to_reserve; int ret; if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) || !is_fstree(root->objectid) || len == 0) return 0; - changeset.bytes_changed = 0; - ulist_init(&changeset.range_changed); + /* @reserved parameter is mandatory for qgroup */ + if (WARN_ON(!reserved_ret)) + return -EINVAL; + if (!*reserved_ret) { + *reserved_ret = extent_changeset_alloc(); + if (!*reserved_ret) + return -ENOMEM; + } + reserved = *reserved_ret; + /* Record already reserved space */ + orig_reserved = reserved->bytes_changed; ret = set_record_extent_bits(&BTRFS_I(inode)->io_tree, start, - start + len -1, EXTENT_QGROUP_RESERVED, &changeset); + start + len -1, EXTENT_QGROUP_RESERVED, reserved); + + /* Newly reserved space */ + to_reserve = reserved->bytes_changed - orig_reserved; trace_btrfs_qgroup_reserve_data(inode, start, len, - changeset.bytes_changed, - QGROUP_RESERVE); + to_reserve, QGROUP_RESERVE); if (ret < 0) goto cleanup; - ret = qgroup_reserve(root, changeset.bytes_changed, true); + ret = qgroup_reserve(root, to_reserve, true); if (ret < 0) goto cleanup; - ulist_release(&changeset.range_changed); return ret; cleanup: - /* cleanup already reserved ranges */ + /* cleanup *ALL* already reserved ranges */ ULIST_ITER_INIT(&uiter); - while ((unode = ulist_next(&changeset.range_changed, &uiter))) + while ((unode = ulist_next(&reserved->range_changed, &uiter))) clear_extent_bit(&BTRFS_I(inode)->io_tree, unode->val, unode->aux, EXTENT_QGROUP_RESERVED, 0, 0, NULL, GFP_NOFS); - ulist_release(&changeset.range_changed); + extent_changeset_release(reserved); return ret; } @@ -2882,8 +2899,7 @@ static int __btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len, int trace_op = QGROUP_RELEASE; int ret; - changeset.bytes_changed = 0; - ulist_init(&changeset.range_changed); + extent_changeset_init(&changeset); ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len -1, EXTENT_QGROUP_RESERVED, &changeset); if (ret < 0) @@ -2899,7 +2915,7 @@ static int __btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len, changeset.bytes_changed); ret = changeset.bytes_changed; out: - ulist_release(&changeset.range_changed); + extent_changeset_release(&changeset); return ret; } @@ -2999,8 +3015,7 @@ void btrfs_qgroup_check_reserved_leak(struct inode *inode) struct ulist_iterator iter; int ret; - changeset.bytes_changed = 0; - ulist_init(&changeset.range_changed); + extent_changeset_init(&changeset); ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, 0, (u64)-1, EXTENT_QGROUP_RESERVED, &changeset); @@ -3017,5 +3032,5 @@ void btrfs_qgroup_check_reserved_leak(struct inode *inode) changeset.bytes_changed); } - ulist_release(&changeset.range_changed); + extent_changeset_release(&changeset); } diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index d11125d6afb9..99408e93eb0d 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -242,7 +242,8 @@ int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid, #endif /* New io_tree based accurate qgroup reserve API */ -int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len); +int btrfs_qgroup_reserve_data(struct inode *inode, + struct extent_changeset **reserved, u64 start, u64 len); int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len); int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index b291d1bebb4c..6407423151ab 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -3093,11 +3093,12 @@ int prealloc_file_extent_cluster(struct inode *inode, u64 prealloc_start = cluster->start - offset; u64 prealloc_end = cluster->end - offset; u64 cur_offset; + struct extent_changeset *data_reserved = NULL; BUG_ON(cluster->start != cluster->boundary[0]); inode_lock(inode); - ret = btrfs_check_data_free_space(inode, prealloc_start, + ret = btrfs_check_data_free_space(inode, &data_reserved, prealloc_start, prealloc_end + 1 - prealloc_start); if (ret) goto out; @@ -3129,6 +3130,7 @@ int prealloc_file_extent_cluster(struct inode *inode, prealloc_end + 1 - cur_offset); out: inode_unlock(inode); + extent_changeset_free(data_reserved); return ret; } -- cgit v1.2.3 From bc42bda22345efdb5d8b578d1b4df2c6eaa85c58 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 27 Feb 2017 15:10:39 +0800 Subject: btrfs: qgroup: Fix qgroup reserved space underflow by only freeing reserved ranges [BUG] For the following case, btrfs can underflow qgroup reserved space at an error path: (Page size 4K, function name without "btrfs_" prefix) Task A | Task B ---------------------------------------------------------------------- Buffered_write [0, 2K) | |- check_data_free_space() | | |- qgroup_reserve_data() | | Range aligned to page | | range [0, 4K) <<< | | 4K bytes reserved <<< | |- copy pages to page cache | | Buffered_write [2K, 4K) | |- check_data_free_space() | | |- qgroup_reserved_data() | | Range alinged to page | | range [0, 4K) | | Already reserved by A <<< | | 0 bytes reserved <<< | |- delalloc_reserve_metadata() | | And it *FAILED* (Maybe EQUOTA) | |- free_reserved_data_space() |- qgroup_free_data() Range aligned to page range [0, 4K) Freeing 4K (Special thanks to Chandan for the detailed report and analyse) [CAUSE] Above Task B is freeing reserved data range [0, 4K) which is actually reserved by Task A. And at writeback time, page dirty by Task A will go through writeback routine, which will free 4K reserved data space at file extent insert time, causing the qgroup underflow. [FIX] For btrfs_qgroup_free_data(), add @reserved parameter to only free data ranges reserved by previous btrfs_qgroup_reserve_data(). So in above case, Task B will try to free 0 byte, so no underflow. Reported-by: Chandan Rajendra Signed-off-by: Qu Wenruo Reviewed-by: Chandan Rajendra Tested-by: Chandan Rajendra Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 6 +++-- fs/btrfs/extent-tree.c | 12 +++++---- fs/btrfs/file.c | 29 +++++++++++--------- fs/btrfs/inode.c | 29 ++++++++++---------- fs/btrfs/ioctl.c | 4 +-- fs/btrfs/qgroup.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++---- fs/btrfs/qgroup.h | 3 ++- fs/btrfs/relocation.c | 8 +++--- 8 files changed, 117 insertions(+), 46 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 1ee4489dc398..5bdd36664421 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2711,7 +2711,10 @@ enum btrfs_flush_state { int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes); int btrfs_check_data_free_space(struct inode *inode, struct extent_changeset **reserved, u64 start, u64 len); -void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len); +void btrfs_free_reserved_data_space(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len); +void btrfs_delalloc_release_space(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len); void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start, u64 len); void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans, @@ -2730,7 +2733,6 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes); void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes); int btrfs_delalloc_reserve_space(struct inode *inode, struct extent_changeset **reserved, u64 start, u64 len); -void btrfs_delalloc_release_space(struct inode *inode, u64 start, u64 len); void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type); struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info, unsigned short type); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 70f85cfdbd46..a3339acec28c 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4389,7 +4389,8 @@ void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start, * This one will handle the per-inode data rsv map for accurate reserved * space framework. */ -void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len) +void btrfs_free_reserved_data_space(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len) { struct btrfs_root *root = BTRFS_I(inode)->root; @@ -4399,7 +4400,7 @@ void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len) start = round_down(start, root->fs_info->sectorsize); btrfs_free_reserved_data_space_noquota(inode, start, len); - btrfs_qgroup_free_data(inode, start, len); + btrfs_qgroup_free_data(inode, reserved, start, len); } static void force_metadata_allocation(struct btrfs_fs_info *info) @@ -6204,7 +6205,7 @@ int btrfs_delalloc_reserve_space(struct inode *inode, return ret; ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len); if (ret < 0) - btrfs_free_reserved_data_space(inode, start, len); + btrfs_free_reserved_data_space(inode, *reserved, start, len); return ret; } @@ -6223,10 +6224,11 @@ int btrfs_delalloc_reserve_space(struct inode *inode, * list if there are no delalloc bytes left. * Also it will handle the qgroup reserved space. */ -void btrfs_delalloc_release_space(struct inode *inode, u64 start, u64 len) +void btrfs_delalloc_release_space(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len) { btrfs_delalloc_release_metadata(BTRFS_I(inode), len); - btrfs_free_reserved_data_space(inode, start, len); + btrfs_free_reserved_data_space(inode, reserved, start, len); } static int update_block_group(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 1b5cce51728b..0f102a1b851f 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1660,8 +1660,9 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, reserve_bytes); if (ret) { if (!only_release_metadata) - btrfs_free_reserved_data_space(inode, pos, - write_bytes); + btrfs_free_reserved_data_space(inode, + data_reserved, pos, + write_bytes); else btrfs_end_write_no_snapshoting(root); break; @@ -1743,8 +1744,9 @@ again: __pos = round_down(pos, fs_info->sectorsize) + (dirty_pages << PAGE_SHIFT); - btrfs_delalloc_release_space(inode, __pos, - release_bytes); + btrfs_delalloc_release_space(inode, + data_reserved, __pos, + release_bytes); } } @@ -1799,9 +1801,9 @@ again: btrfs_delalloc_release_metadata(BTRFS_I(inode), release_bytes); } else { - btrfs_delalloc_release_space(inode, - round_down(pos, fs_info->sectorsize), - release_bytes); + btrfs_delalloc_release_space(inode, data_reserved, + round_down(pos, fs_info->sectorsize), + release_bytes); } } @@ -2918,8 +2920,8 @@ static long btrfs_fallocate(struct file *file, int mode, * range, free reserved data space first, otherwise * it'll result in false ENOSPC error. */ - btrfs_free_reserved_data_space(inode, cur_offset, - last_byte - cur_offset); + btrfs_free_reserved_data_space(inode, data_reserved, + cur_offset, last_byte - cur_offset); } free_extent_map(em); cur_offset = last_byte; @@ -2938,8 +2940,9 @@ static long btrfs_fallocate(struct file *file, int mode, range->len, i_blocksize(inode), offset + len, &alloc_hint); else - btrfs_free_reserved_data_space(inode, range->start, - range->len); + btrfs_free_reserved_data_space(inode, + data_reserved, range->start, + range->len); list_del(&range->list); kfree(range); } @@ -2977,8 +2980,8 @@ out: inode_unlock(inode); /* Let go of our reservation. */ if (ret != 0) - btrfs_free_reserved_data_space(inode, alloc_start, - alloc_end - cur_offset); + btrfs_free_reserved_data_space(inode, data_reserved, + alloc_start, alloc_end - cur_offset); extent_changeset_free(data_reserved); return ret; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d9cf6df40d5e..5d3c6ac960fd 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -345,7 +345,7 @@ out: * And at reserve time, it's always aligned to page size, so * just free one page here. */ - btrfs_qgroup_free_data(inode, 0, PAGE_SIZE); + btrfs_qgroup_free_data(inode, NULL, 0, PAGE_SIZE); btrfs_free_path(path); btrfs_end_transaction(trans); return ret; @@ -2935,7 +2935,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) * space for NOCOW range. * As NOCOW won't cause a new delayed ref, just free the space */ - btrfs_qgroup_free_data(inode, ordered_extent->file_offset, + btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset, ordered_extent->len); btrfs_ordered_update_i_size(inode, 0, ordered_extent); if (nolock) @@ -4794,7 +4794,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, again: page = find_or_create_page(mapping, index, mask); if (!page) { - btrfs_delalloc_release_space(inode, + btrfs_delalloc_release_space(inode, data_reserved, round_down(from, blocksize), blocksize); ret = -ENOMEM; @@ -4866,7 +4866,7 @@ again: out_unlock: if (ret) - btrfs_delalloc_release_space(inode, block_start, + btrfs_delalloc_release_space(inode, data_reserved, block_start, blocksize); unlock_page(page); put_page(page); @@ -5266,7 +5266,7 @@ static void evict_inode_truncate_pages(struct inode *inode) * Note, end is the bytenr of last byte, so we need + 1 here. */ if (state->state & EXTENT_DELALLOC) - btrfs_qgroup_free_data(inode, start, end - start + 1); + btrfs_qgroup_free_data(inode, NULL, start, end - start + 1); clear_extent_bit(io_tree, start, end, EXTENT_LOCKED | EXTENT_DIRTY | @@ -8792,8 +8792,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) current->journal_info = NULL; if (ret < 0 && ret != -EIOCBQUEUED) { if (dio_data.reserve) - btrfs_delalloc_release_space(inode, offset, - dio_data.reserve); + btrfs_delalloc_release_space(inode, data_reserved, + offset, dio_data.reserve); /* * On error we might have left some ordered extents * without submitting corresponding bios for them, so @@ -8808,8 +8808,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) dio_data.unsubmitted_oe_range_start, false); } else if (ret >= 0 && (size_t)ret < count) - btrfs_delalloc_release_space(inode, offset, - count - (size_t)ret); + btrfs_delalloc_release_space(inode, data_reserved, + offset, count - (size_t)ret); } out: if (wakeup) @@ -9008,7 +9008,7 @@ again: * free the entire extent. */ if (PageDirty(page)) - btrfs_qgroup_free_data(inode, page_start, PAGE_SIZE); + btrfs_qgroup_free_data(inode, NULL, page_start, PAGE_SIZE); if (!inode_evicting) { clear_extent_bit(tree, page_start, page_end, EXTENT_LOCKED | EXTENT_DIRTY | @@ -9130,8 +9130,8 @@ again: spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents++; spin_unlock(&BTRFS_I(inode)->lock); - btrfs_delalloc_release_space(inode, page_start, - PAGE_SIZE - reserved_space); + btrfs_delalloc_release_space(inode, data_reserved, + page_start, PAGE_SIZE - reserved_space); } } @@ -9187,7 +9187,8 @@ out_unlock: } unlock_page(page); out: - btrfs_delalloc_release_space(inode, page_start, reserved_space); + btrfs_delalloc_release_space(inode, data_reserved, page_start, + reserved_space); out_noreserve: sb_end_pagefault(inode->i_sb); extent_changeset_free(data_reserved); @@ -10557,7 +10558,7 @@ next: btrfs_end_transaction(trans); } if (cur_offset < end) - btrfs_free_reserved_data_space(inode, cur_offset, + btrfs_free_reserved_data_space(inode, NULL, cur_offset, end - cur_offset + 1); return ret; } diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ccee5417d3f6..b4e9941efb60 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1227,7 +1227,7 @@ again: spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents++; spin_unlock(&BTRFS_I(inode)->lock); - btrfs_delalloc_release_space(inode, + btrfs_delalloc_release_space(inode, data_reserved, start_index << PAGE_SHIFT, (page_cnt - i_done) << PAGE_SHIFT); } @@ -1255,7 +1255,7 @@ out: unlock_page(pages[i]); put_page(pages[i]); } - btrfs_delalloc_release_space(inode, + btrfs_delalloc_release_space(inode, data_reserved, start_index << PAGE_SHIFT, page_cnt << PAGE_SHIFT); extent_changeset_free(data_reserved); diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 002088937021..fc9dffaa9524 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -2892,13 +2892,72 @@ cleanup: return ret; } -static int __btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len, - int free) +/* Free ranges specified by @reserved, normally in error path */ +static int qgroup_free_reserved_data(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len) +{ + struct btrfs_root *root = BTRFS_I(inode)->root; + struct ulist_node *unode; + struct ulist_iterator uiter; + struct extent_changeset changeset; + int freed = 0; + int ret; + + extent_changeset_init(&changeset); + len = round_up(start + len, root->fs_info->sectorsize); + start = round_down(start, root->fs_info->sectorsize); + + ULIST_ITER_INIT(&uiter); + while ((unode = ulist_next(&reserved->range_changed, &uiter))) { + u64 range_start = unode->val; + /* unode->aux is the inclusive end */ + u64 range_len = unode->aux - range_start + 1; + u64 free_start; + u64 free_len; + + extent_changeset_release(&changeset); + + /* Only free range in range [start, start + len) */ + if (range_start >= start + len || + range_start + range_len <= start) + continue; + free_start = max(range_start, start); + free_len = min(start + len, range_start + range_len) - + free_start; + /* + * TODO: To also modify reserved->ranges_reserved to reflect + * the modification. + * + * However as long as we free qgroup reserved according to + * EXTENT_QGROUP_RESERVED, we won't double free. + * So not need to rush. + */ + ret = clear_record_extent_bits(&BTRFS_I(inode)->io_failure_tree, + free_start, free_start + free_len - 1, + EXTENT_QGROUP_RESERVED, &changeset); + if (ret < 0) + goto out; + freed += changeset.bytes_changed; + } + btrfs_qgroup_free_refroot(root->fs_info, root->objectid, freed); + ret = freed; +out: + extent_changeset_release(&changeset); + return ret; +} + +static int __btrfs_qgroup_release_data(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len, + int free) { struct extent_changeset changeset; int trace_op = QGROUP_RELEASE; int ret; + /* In release case, we shouldn't have @reserved */ + WARN_ON(!free && reserved); + if (free && reserved) + return qgroup_free_reserved_data(inode, reserved, start, len); extent_changeset_init(&changeset); ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len -1, EXTENT_QGROUP_RESERVED, &changeset); @@ -2924,14 +2983,17 @@ out: * * Should be called when a range of pages get invalidated before reaching disk. * Or for error cleanup case. + * if @reserved is given, only reserved range in [@start, @start + @len) will + * be freed. * * For data written to disk, use btrfs_qgroup_release_data(). * * NOTE: This function may sleep for memory allocation. */ -int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len) +int btrfs_qgroup_free_data(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len) { - return __btrfs_qgroup_release_data(inode, start, len, 1); + return __btrfs_qgroup_release_data(inode, reserved, start, len, 1); } /* @@ -2951,7 +3013,7 @@ int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len) */ int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len) { - return __btrfs_qgroup_release_data(inode, start, len, 0); + return __btrfs_qgroup_release_data(inode, NULL, start, len, 0); } int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index 99408e93eb0d..d9984e87cddf 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -245,7 +245,8 @@ int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid, int btrfs_qgroup_reserve_data(struct inode *inode, struct extent_changeset **reserved, u64 start, u64 len); int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len); -int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len); +int btrfs_qgroup_free_data(struct inode *inode, + struct extent_changeset *reserved, u64 start, u64 len); int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, bool enforce); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 6407423151ab..dc69b6ba29af 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -3114,8 +3114,8 @@ int prealloc_file_extent_cluster(struct inode *inode, lock_extent(&BTRFS_I(inode)->io_tree, start, end); num_bytes = end + 1 - start; if (cur_offset < start) - btrfs_free_reserved_data_space(inode, cur_offset, - start - cur_offset); + btrfs_free_reserved_data_space(inode, data_reserved, + cur_offset, start - cur_offset); ret = btrfs_prealloc_file_range(inode, 0, start, num_bytes, num_bytes, end + 1, &alloc_hint); @@ -3126,8 +3126,8 @@ int prealloc_file_extent_cluster(struct inode *inode, nr++; } if (cur_offset < prealloc_end) - btrfs_free_reserved_data_space(inode, cur_offset, - prealloc_end + 1 - cur_offset); + btrfs_free_reserved_data_space(inode, data_reserved, + cur_offset, prealloc_end + 1 - cur_offset); out: inode_unlock(inode); extent_changeset_free(data_reserved); -- cgit v1.2.3