--- zzzz-none-000/linux-3.10.107/fs/btrfs/extent_io.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/fs/btrfs/extent_io.c 2021-02-04 17:41:59.000000000 +0000 @@ -13,18 +13,23 @@ #include #include "extent_io.h" #include "extent_map.h" -#include "compat.h" #include "ctree.h" #include "btrfs_inode.h" #include "volumes.h" #include "check-integrity.h" #include "locking.h" #include "rcu-string.h" +#include "backref.h" static struct kmem_cache *extent_state_cache; static struct kmem_cache *extent_buffer_cache; static struct bio_set *btrfs_bioset; +static inline bool extent_state_in_tree(const struct extent_state *state) +{ + return !RB_EMPTY_NODE(&state->rb_node); +} + #ifdef CONFIG_BTRFS_DEBUG static LIST_HEAD(buffers); static LIST_HEAD(states); @@ -59,28 +64,48 @@ while (!list_empty(&states)) { state = list_entry(states.next, struct extent_state, leak_list); - printk(KERN_ERR "btrfs state leak: start %llu end %llu " - "state %lu in tree %p refs %d\n", - (unsigned long long)state->start, - (unsigned long long)state->end, - state->state, state->tree, atomic_read(&state->refs)); + pr_err("BTRFS: state leak: start %llu end %llu state %u in tree %d refs %d\n", + state->start, state->end, state->state, + extent_state_in_tree(state), + atomic_read(&state->refs)); list_del(&state->leak_list); kmem_cache_free(extent_state_cache, state); } while (!list_empty(&buffers)) { eb = list_entry(buffers.next, struct extent_buffer, leak_list); - printk(KERN_ERR "btrfs buffer leak start %llu len %lu " - "refs %d\n", (unsigned long long)eb->start, - eb->len, atomic_read(&eb->refs)); + printk(KERN_ERR "BTRFS: buffer leak start %llu len %lu " + "refs %d\n", + eb->start, eb->len, atomic_read(&eb->refs)); list_del(&eb->leak_list); kmem_cache_free(extent_buffer_cache, eb); } } + +#define btrfs_debug_check_extent_io_range(tree, start, end) \ + __btrfs_debug_check_extent_io_range(__func__, (tree), (start), (end)) +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(inode), isize, start, end); + } +} #else #define btrfs_leak_debug_add(new, head) do {} while (0) #define btrfs_leak_debug_del(entry) do {} while (0) #define btrfs_leak_debug_check() do {} while (0) +#define btrfs_debug_check_extent_io_range(c, s, e) do {} while (0) #endif #define BUFFER_LRU_MAX 64 @@ -106,10 +131,31 @@ unsigned int sync_io:1; }; +static void add_extent_changeset(struct extent_state *state, unsigned bits, + struct extent_changeset *changeset, + int set) +{ + int ret; + + if (!changeset) + return; + if (set && (state->state & bits) == bits) + return; + if (!set && (state->state & bits) == 0) + return; + changeset->bytes_changed += state->end - state->start + 1; + ret = ulist_add(changeset->range_changed, state->start, state->end, + GFP_ATOMIC); + /* ENOMEM */ + BUG_ON(ret < 0); +} + 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); } @@ -131,8 +177,16 @@ offsetof(struct btrfs_io_bio, bio)); if (!btrfs_bioset) goto free_buffer_cache; + + if (bioset_integrity_create(btrfs_bioset, BIO_POOL_SIZE)) + goto free_bioset; + return 0; +free_bioset: + bioset_free(btrfs_bioset); + btrfs_bioset = NULL; + free_buffer_cache: kmem_cache_destroy(extent_buffer_cache); extent_buffer_cache = NULL; @@ -164,11 +218,9 @@ struct address_space *mapping) { tree->state = RB_ROOT; - INIT_RADIX_TREE(&tree->buffer, GFP_ATOMIC); tree->ops = NULL; tree->dirty_bytes = 0; spin_lock_init(&tree->lock); - spin_lock_init(&tree->buffer_lock); tree->mapping = mapping; } @@ -181,7 +233,7 @@ return state; state->state = 0; state->private = 0; - state->tree = NULL; + RB_CLEAR_NODE(&state->rb_node); btrfs_leak_debug_add(&state->leak_list, &states); atomic_set(&state->refs, 1); init_waitqueue_head(&state->wq); @@ -194,20 +246,31 @@ if (!state) return; if (atomic_dec_and_test(&state->refs)) { - WARN_ON(state->tree); + WARN_ON(extent_state_in_tree(state)); btrfs_leak_debug_del(&state->leak_list); trace_free_extent_state(state, _RET_IP_); kmem_cache_free(extent_state_cache, state); } } -static struct rb_node *tree_insert(struct rb_root *root, u64 offset, - struct rb_node *node) +static struct rb_node *tree_insert(struct rb_root *root, + struct rb_node *search_start, + u64 offset, + struct rb_node *node, + struct rb_node ***p_in, + struct rb_node **parent_in) { - struct rb_node **p = &root->rb_node; + struct rb_node **p; struct rb_node *parent = NULL; struct tree_entry *entry; + if (p_in && parent_in) { + p = *p_in; + parent = *parent_in; + goto do_insert; + } + + p = search_start ? &search_start : &root->rb_node; while (*p) { parent = *p; entry = rb_entry(parent, struct tree_entry, rb_node); @@ -220,35 +283,43 @@ return parent; } +do_insert: rb_link_node(node, parent, p); rb_insert_color(node, root); return NULL; } static struct rb_node *__etree_search(struct extent_io_tree *tree, u64 offset, - struct rb_node **prev_ret, - struct rb_node **next_ret) + struct rb_node **prev_ret, + struct rb_node **next_ret, + struct rb_node ***p_ret, + struct rb_node **parent_ret) { struct rb_root *root = &tree->state; - struct rb_node *n = root->rb_node; + struct rb_node **n = &root->rb_node; struct rb_node *prev = NULL; struct rb_node *orig_prev = NULL; struct tree_entry *entry; struct tree_entry *prev_entry = NULL; - while (n) { - entry = rb_entry(n, struct tree_entry, rb_node); - prev = n; + while (*n) { + prev = *n; + entry = rb_entry(prev, struct tree_entry, rb_node); prev_entry = entry; if (offset < entry->start) - n = n->rb_left; + n = &(*n)->rb_left; else if (offset > entry->end) - n = n->rb_right; + n = &(*n)->rb_right; else - return n; + return *n; } + if (p_ret) + *p_ret = n; + if (parent_ret) + *parent_ret = prev; + if (prev_ret) { orig_prev = prev; while (prev && offset > prev_entry->end) { @@ -270,18 +341,27 @@ return NULL; } -static inline struct rb_node *tree_search(struct extent_io_tree *tree, - u64 offset) +static inline struct rb_node * +tree_search_for_insert(struct extent_io_tree *tree, + u64 offset, + struct rb_node ***p_ret, + struct rb_node **parent_ret) { struct rb_node *prev = NULL; struct rb_node *ret; - ret = __etree_search(tree, offset, &prev, NULL); + ret = __etree_search(tree, offset, &prev, NULL, p_ret, parent_ret); if (!ret) return prev; return ret; } +static inline struct rb_node *tree_search(struct extent_io_tree *tree, + u64 offset) +{ + return tree_search_for_insert(tree, offset, NULL, NULL); +} + static void merge_cb(struct extent_io_tree *tree, struct extent_state *new, struct extent_state *other) { @@ -315,8 +395,8 @@ other->state == state->state) { merge_cb(tree, state, other); state->start = other->start; - other->tree = NULL; rb_erase(&other->rb_node, &tree->state); + RB_CLEAR_NODE(&other->rb_node); free_extent_state(other); } } @@ -327,29 +407,30 @@ other->state == state->state) { merge_cb(tree, state, other); state->end = other->end; - other->tree = NULL; rb_erase(&other->rb_node, &tree->state); + RB_CLEAR_NODE(&other->rb_node); free_extent_state(other); } } } static void set_state_cb(struct extent_io_tree *tree, - struct extent_state *state, unsigned long *bits) + struct extent_state *state, unsigned *bits) { if (tree->ops && tree->ops->set_bit_hook) tree->ops->set_bit_hook(tree->mapping->host, state, bits); } static void clear_state_cb(struct extent_io_tree *tree, - struct extent_state *state, unsigned long *bits) + struct extent_state *state, unsigned *bits) { if (tree->ops && tree->ops->clear_bit_hook) tree->ops->clear_bit_hook(tree->mapping->host, state, bits); } static void set_state_bits(struct extent_io_tree *tree, - struct extent_state *state, unsigned long *bits); + struct extent_state *state, unsigned *bits, + struct extent_changeset *changeset); /* * insert an extent_state struct into the tree. 'bits' are set on the @@ -363,30 +444,29 @@ */ static int insert_state(struct extent_io_tree *tree, struct extent_state *state, u64 start, u64 end, - unsigned long *bits) + struct rb_node ***p, + struct rb_node **parent, + unsigned *bits, struct extent_changeset *changeset) { struct rb_node *node; if (end < start) - WARN(1, KERN_ERR "btrfs end < start %llu %llu\n", - (unsigned long long)end, - (unsigned long long)start); + WARN(1, KERN_ERR "BTRFS: end < start %llu %llu\n", + end, start); state->start = start; state->end = end; - set_state_bits(tree, state, bits); + set_state_bits(tree, state, bits, changeset); - node = tree_insert(&tree->state, end, &state->rb_node); + node = tree_insert(&tree->state, NULL, end, &state->rb_node, p, parent); if (node) { struct extent_state *found; found = rb_entry(node, struct extent_state, rb_node); - printk(KERN_ERR "btrfs found node %llu %llu on insert of " - "%llu %llu\n", (unsigned long long)found->start, - (unsigned long long)found->end, - (unsigned long long)start, (unsigned long long)end); + printk(KERN_ERR "BTRFS: found node %llu %llu on insert of " + "%llu %llu\n", + found->start, found->end, start, end); return -EEXIST; } - state->tree = tree; merge_state(tree, state); return 0; } @@ -424,12 +504,12 @@ prealloc->state = orig->state; orig->start = split; - node = tree_insert(&tree->state, prealloc->end, &prealloc->rb_node); + node = tree_insert(&tree->state, &orig->rb_node, prealloc->end, + &prealloc->rb_node, NULL, NULL); if (node) { free_extent_state(prealloc); return -EEXIST; } - prealloc->tree = tree; return 0; } @@ -451,10 +531,11 @@ */ static struct extent_state *clear_state_bit(struct extent_io_tree *tree, struct extent_state *state, - unsigned long *bits, int wake) + unsigned *bits, int wake, + struct extent_changeset *changeset) { struct extent_state *next; - unsigned long bits_to_clear = *bits & ~EXTENT_CTLBITS; + unsigned bits_to_clear = *bits & ~EXTENT_CTLBITS; if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) { u64 range = state->end - state->start + 1; @@ -462,14 +543,15 @@ tree->dirty_bytes -= range; } clear_state_cb(tree, state, bits); + add_extent_changeset(state, bits_to_clear, changeset, 0); state->state &= ~bits_to_clear; if (wake) wake_up(&state->wq); if (state->state == 0) { next = next_state(state); - if (state->tree) { + if (extent_state_in_tree(state)) { rb_erase(&state->rb_node, &tree->state); - state->tree = NULL; + RB_CLEAR_NODE(&state->rb_node); free_extent_state(state); } else { WARN_ON(1); @@ -509,10 +591,10 @@ * * This takes the tree lock, and returns 0 on success and < 0 on error. */ -int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, int wake, int delete, - struct extent_state **cached_state, - gfp_t mask) +static int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, + unsigned bits, int wake, int delete, + struct extent_state **cached_state, + gfp_t mask, struct extent_changeset *changeset) { struct extent_state *state; struct extent_state *cached; @@ -522,6 +604,11 @@ int err; int clear = 0; + btrfs_debug_check_extent_io_range(tree, start, end); + + if (bits & EXTENT_DELALLOC) + bits |= EXTENT_NORESERVE; + if (delete) bits |= ~EXTENT_CTLBITS; bits |= EXTENT_FIRST_DELALLOC; @@ -529,10 +616,15 @@ if (bits & (EXTENT_IOBITS | EXTENT_BOUNDARY)) clear = 1; again: - if (!prealloc && (mask & __GFP_WAIT)) { + if (!prealloc && gfpflags_allow_blocking(mask)) { + /* + * Don't care for allocation failure here because we might end + * up not needing the pre-allocated extent state at all, which + * is the case if we only have in the tree extent states that + * cover our input range and don't cover too any other range. + * If we end up needing a new extent state we allocate it later. + */ prealloc = alloc_extent_state(mask); - if (!prealloc) - return -ENOMEM; } spin_lock(&tree->lock); @@ -544,8 +636,8 @@ cached_state = NULL; } - if (cached && cached->tree && cached->start <= start && - cached->end > start) { + if (cached && extent_state_in_tree(cached) && + cached->start <= start && cached->end > start) { if (clear) atomic_dec(&cached->refs); state = cached; @@ -601,7 +693,8 @@ if (err) goto out; if (state->end <= end) { - state = clear_state_bit(tree, state, &bits, wake); + state = clear_state_bit(tree, state, &bits, wake, + changeset); goto next; } goto search_again; @@ -622,13 +715,13 @@ if (wake) wake_up(&state->wq); - clear_state_bit(tree, prealloc, &bits, wake); + clear_state_bit(tree, prealloc, &bits, wake, changeset); prealloc = NULL; goto out; } - state = clear_state_bit(tree, state, &bits, wake); + state = clear_state_bit(tree, state, &bits, wake, changeset); next: if (last_end == (u64)-1) goto out; @@ -648,7 +741,7 @@ if (start > end) goto out; spin_unlock(&tree->lock); - if (mask & __GFP_WAIT) + if (gfpflags_allow_blocking(mask)) cond_resched(); goto again; } @@ -677,6 +770,8 @@ struct extent_state *state; struct rb_node *node; + btrfs_debug_check_extent_io_range(tree, start, end); + spin_lock(&tree->lock); again: while (1) { @@ -685,6 +780,7 @@ * our range starts */ node = tree_search(tree, start); +process_node: if (!node) break; @@ -705,7 +801,10 @@ if (start > end) break; - cond_resched_lock(&tree->lock); + if (!cond_resched_lock(&tree->lock)) { + node = rb_next(node); + goto process_node; + } } out: spin_unlock(&tree->lock); @@ -713,36 +812,36 @@ static void set_state_bits(struct extent_io_tree *tree, struct extent_state *state, - unsigned long *bits) + unsigned *bits, struct extent_changeset *changeset) { - unsigned long bits_to_set = *bits & ~EXTENT_CTLBITS; + unsigned bits_to_set = *bits & ~EXTENT_CTLBITS; set_state_cb(tree, state, bits); if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) { u64 range = state->end - state->start + 1; tree->dirty_bytes += range; } + add_extent_changeset(state, bits_to_set, changeset, 1); state->state |= bits_to_set; } -static void cache_state(struct extent_state *state, - struct extent_state **cached_ptr) +static void cache_state_if_flags(struct extent_state *state, + struct extent_state **cached_ptr, + unsigned flags) { if (cached_ptr && !(*cached_ptr)) { - if (state->state & (EXTENT_IOBITS | EXTENT_BOUNDARY)) { + if (!flags || (state->state & flags)) { *cached_ptr = state; atomic_inc(&state->refs); } } } -static void uncache_state(struct extent_state **cached_ptr) +static void cache_state(struct extent_state *state, + struct extent_state **cached_ptr) { - if (cached_ptr && (*cached_ptr)) { - struct extent_state *state = *cached_ptr; - *cached_ptr = NULL; - free_extent_state(state); - } + return cache_state_if_flags(state, cached_ptr, + EXTENT_IOBITS | EXTENT_BOUNDARY); } /* @@ -758,20 +857,24 @@ static int __must_check __set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, unsigned long exclusive_bits, + unsigned bits, unsigned exclusive_bits, u64 *failed_start, struct extent_state **cached_state, - gfp_t mask) + gfp_t mask, struct extent_changeset *changeset) { struct extent_state *state; struct extent_state *prealloc = NULL; struct rb_node *node; + struct rb_node **p; + struct rb_node *parent; int err = 0; u64 last_start; u64 last_end; + btrfs_debug_check_extent_io_range(tree, start, end); + bits |= EXTENT_FIRST_DELALLOC; again: - if (!prealloc && (mask & __GFP_WAIT)) { + if (!prealloc && gfpflags_allow_blocking(mask)) { prealloc = alloc_extent_state(mask); BUG_ON(!prealloc); } @@ -780,7 +883,7 @@ if (cached_state && *cached_state) { state = *cached_state; if (state->start <= start && state->end > start && - state->tree) { + extent_state_in_tree(state)) { node = &state->rb_node; goto hit_next; } @@ -789,14 +892,16 @@ * this search will find all the extents that end after * our range starts. */ - node = tree_search(tree, start); + node = tree_search_for_insert(tree, start, &p, &parent); if (!node) { prealloc = alloc_extent_state_atomic(prealloc); BUG_ON(!prealloc); - err = insert_state(tree, prealloc, start, end, &bits); + err = insert_state(tree, prealloc, start, end, + &p, &parent, &bits, changeset); if (err) extent_io_tree_panic(tree, err); + cache_state(prealloc, cached_state); prealloc = NULL; goto out; } @@ -818,7 +923,7 @@ goto out; } - set_state_bits(tree, state, &bits); + set_state_bits(tree, state, &bits, changeset); cache_state(state, cached_state); merge_state(tree, state); if (last_end == (u64)-1) @@ -864,7 +969,7 @@ if (err) goto out; if (state->end <= end) { - set_state_bits(tree, state, &bits); + set_state_bits(tree, state, &bits, changeset); cache_state(state, cached_state); merge_state(tree, state); if (last_end == (u64)-1) @@ -899,7 +1004,7 @@ * the later extent. */ err = insert_state(tree, prealloc, start, this_end, - &bits); + NULL, NULL, &bits, changeset); if (err) extent_io_tree_panic(tree, err); @@ -927,7 +1032,7 @@ if (err) extent_io_tree_panic(tree, err); - set_state_bits(tree, prealloc, &bits); + set_state_bits(tree, prealloc, &bits, changeset); cache_state(prealloc, cached_state); merge_state(tree, prealloc); prealloc = NULL; @@ -947,17 +1052,17 @@ if (start > end) goto out; spin_unlock(&tree->lock); - if (mask & __GFP_WAIT) + if (gfpflags_allow_blocking(mask)) cond_resched(); goto again; } int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, u64 * failed_start, + unsigned bits, u64 * failed_start, struct extent_state **cached_state, gfp_t mask) { return __set_extent_bit(tree, start, end, bits, 0, failed_start, - cached_state, mask); + cached_state, mask, NULL); } @@ -979,20 +1084,32 @@ * boundary bits like LOCK. */ int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, unsigned long clear_bits, + unsigned bits, unsigned clear_bits, struct extent_state **cached_state, gfp_t mask) { struct extent_state *state; struct extent_state *prealloc = NULL; struct rb_node *node; + struct rb_node **p; + struct rb_node *parent; int err = 0; u64 last_start; u64 last_end; + bool first_iteration = true; + + btrfs_debug_check_extent_io_range(tree, start, end); again: - if (!prealloc && (mask & __GFP_WAIT)) { + if (!prealloc && gfpflags_allow_blocking(mask)) { + /* + * Best effort, don't worry if extent state allocation fails + * here for the first iteration. We might have a cached state + * that matches exactly the target range, in which case no + * extent state allocations are needed. We'll only know this + * after locking the tree. + */ prealloc = alloc_extent_state(mask); - if (!prealloc) + if (!prealloc && !first_iteration) return -ENOMEM; } @@ -1000,7 +1117,7 @@ if (cached_state && *cached_state) { state = *cached_state; if (state->start <= start && state->end > start && - state->tree) { + extent_state_in_tree(state)) { node = &state->rb_node; goto hit_next; } @@ -1010,17 +1127,19 @@ * this search will find all the extents that end after * our range starts. */ - node = tree_search(tree, start); + node = tree_search_for_insert(tree, start, &p, &parent); if (!node) { prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) { err = -ENOMEM; goto out; } - err = insert_state(tree, prealloc, start, end, &bits); - prealloc = NULL; + err = insert_state(tree, prealloc, start, end, + &p, &parent, &bits, NULL); if (err) extent_io_tree_panic(tree, err); + cache_state(prealloc, cached_state); + prealloc = NULL; goto out; } state = rb_entry(node, struct extent_state, rb_node); @@ -1035,9 +1154,9 @@ * Just lock what we found and keep going */ if (state->start == start && state->end <= end) { - set_state_bits(tree, state, &bits); + set_state_bits(tree, state, &bits, NULL); cache_state(state, cached_state); - state = clear_state_bit(tree, state, &clear_bits, 0); + state = clear_state_bit(tree, state, &clear_bits, 0, NULL); if (last_end == (u64)-1) goto out; start = last_end + 1; @@ -1076,9 +1195,10 @@ if (err) goto out; if (state->end <= end) { - set_state_bits(tree, state, &bits); + set_state_bits(tree, state, &bits, NULL); cache_state(state, cached_state); - state = clear_state_bit(tree, state, &clear_bits, 0); + state = clear_state_bit(tree, state, &clear_bits, 0, + NULL); if (last_end == (u64)-1) goto out; start = last_end + 1; @@ -1113,7 +1233,7 @@ * the later extent. */ err = insert_state(tree, prealloc, start, this_end, - &bits); + NULL, NULL, &bits, NULL); if (err) extent_io_tree_panic(tree, err); cache_state(prealloc, cached_state); @@ -1138,9 +1258,9 @@ if (err) extent_io_tree_panic(tree, err); - set_state_bits(tree, prealloc, &bits); + set_state_bits(tree, prealloc, &bits, NULL); cache_state(prealloc, cached_state); - clear_state_bit(tree, prealloc, &clear_bits, 0); + clear_state_bit(tree, prealloc, &clear_bits, 0, NULL); prealloc = NULL; goto out; } @@ -1158,8 +1278,9 @@ if (start > end) goto out; spin_unlock(&tree->lock); - if (mask & __GFP_WAIT) + if (gfpflags_allow_blocking(mask)) cond_resched(); + first_iteration = false; goto again; } @@ -1172,16 +1293,59 @@ } int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, gfp_t mask) + unsigned bits, gfp_t mask) { return set_extent_bit(tree, start, end, bits, NULL, NULL, mask); } +int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, + unsigned bits, gfp_t mask, + struct extent_changeset *changeset) +{ + /* + * We don't support EXTENT_LOCKED yet, as current changeset will + * record any bits changed, so for EXTENT_LOCKED case, it will + * either fail with -EEXIST or changeset will record the whole + * range. + */ + BUG_ON(bits & EXTENT_LOCKED); + + return __set_extent_bit(tree, start, end, bits, 0, NULL, NULL, mask, + changeset); +} + +int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, + unsigned bits, int wake, int delete, + struct extent_state **cached, gfp_t mask) +{ + return __clear_extent_bit(tree, start, end, bits, wake, delete, + cached, mask, NULL); +} + int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, gfp_t mask) + unsigned bits, gfp_t mask) { - return clear_extent_bit(tree, start, end, bits, 0, 0, NULL, mask); + int wake = 0; + + if (bits & EXTENT_LOCKED) + wake = 1; + + return clear_extent_bit(tree, start, end, bits, wake, 0, NULL, mask); +} + +int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, + unsigned bits, gfp_t mask, + struct extent_changeset *changeset) +{ + /* + * Don't support EXTENT_LOCKED case, same reason as + * set_record_extent_bits(). + */ + BUG_ON(bits & EXTENT_LOCKED); + + return __clear_extent_bit(tree, start, end, bits, 0, 0, NULL, mask, + changeset); } int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end, @@ -1234,14 +1398,15 @@ * us if waiting is desired. */ int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, struct extent_state **cached_state) + unsigned bits, struct extent_state **cached_state) { int err; u64 failed_start; + while (1) { err = __set_extent_bit(tree, start, end, EXTENT_LOCKED | bits, EXTENT_LOCKED, &failed_start, - cached_state, GFP_NOFS); + cached_state, GFP_NOFS, NULL); if (err == -EEXIST) { wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED); start = failed_start; @@ -1263,7 +1428,7 @@ u64 failed_start; err = __set_extent_bit(tree, start, end, EXTENT_LOCKED, EXTENT_LOCKED, - &failed_start, NULL, GFP_NOFS); + &failed_start, NULL, GFP_NOFS, NULL); if (err == -EEXIST) { if (failed_start > start) clear_extent_bit(tree, start, failed_start - 1, @@ -1311,8 +1476,8 @@ while (index <= end_index) { page = find_get_page(inode->i_mapping, index); BUG_ON(!page); /* Pages should be in the extent_io_tree */ - account_page_redirty(page); __set_page_dirty_nobuffers(page); + account_page_redirty(page); page_cache_release(page); index++; } @@ -1344,7 +1509,7 @@ */ static struct extent_state * find_first_extent_bit_state(struct extent_io_tree *tree, - u64 start, unsigned long bits) + u64 start, unsigned bits) { struct rb_node *node; struct extent_state *state; @@ -1378,7 +1543,7 @@ * If nothing was found, 1 is returned. If found something, return 0. */ int find_first_extent_bit(struct extent_io_tree *tree, u64 start, - u64 *start_ret, u64 *end_ret, unsigned long bits, + u64 *start_ret, u64 *end_ret, unsigned bits, struct extent_state **cached_state) { struct extent_state *state; @@ -1388,7 +1553,7 @@ spin_lock(&tree->lock); if (cached_state && *cached_state) { state = *cached_state; - if (state->end == start - 1 && state->tree) { + if (state->end == start - 1 && extent_state_in_tree(state)) { n = rb_next(&state->rb_node); while (n) { state = rb_entry(n, struct extent_state, @@ -1408,7 +1573,7 @@ state = find_first_extent_bit_state(tree, start, bits); got_it: if (state) { - cache_state(state, cached_state); + cache_state_if_flags(state, cached_state, 0); *start_ret = state->start; *end_ret = state->end; ret = 0; @@ -1467,11 +1632,11 @@ *end = state->end; cur_start = state->end + 1; node = rb_next(node); - if (!node) - break; total_bytes += state->end - state->start + 1; if (total_bytes >= max_bytes) break; + if (!node) + break; } out: spin_unlock(&tree->lock); @@ -1575,11 +1740,10 @@ * * 1 is returned if we find something, 0 if nothing was in the tree */ -static noinline u64 find_lock_delalloc_range(struct inode *inode, - struct extent_io_tree *tree, - struct page *locked_page, - u64 *start, u64 *end, - u64 max_bytes) +STATIC u64 find_lock_delalloc_range(struct inode *inode, + struct extent_io_tree *tree, + struct page *locked_page, u64 *start, + u64 *end, u64 max_bytes) { u64 delalloc_start; u64 delalloc_end; @@ -1598,7 +1762,7 @@ *start = delalloc_start; *end = delalloc_end; free_extent_state(cached_state); - return found; + return 0; } /* @@ -1611,10 +1775,9 @@ /* * make sure to limit the number of pages we try to lock down - * if we're looping. */ - if (delalloc_end + 1 - delalloc_start > max_bytes && loops) - delalloc_end = delalloc_start + PAGE_CACHE_SIZE - 1; + if (delalloc_end + 1 - delalloc_start > max_bytes) + delalloc_end = delalloc_start + max_bytes - 1; /* step two, lock all the pages after the page that has start */ ret = lock_delalloc_pages(inode, locked_page, @@ -1626,8 +1789,7 @@ free_extent_state(cached_state); cached_state = NULL; if (!loops) { - unsigned long offset = (*start) & (PAGE_CACHE_SIZE - 1); - max_bytes = PAGE_CACHE_SIZE - offset; + max_bytes = PAGE_CACHE_SIZE; loops = 1; goto again; } else { @@ -1658,53 +1820,48 @@ return found; } -int extent_clear_unlock_delalloc(struct inode *inode, - struct extent_io_tree *tree, - u64 start, u64 end, struct page *locked_page, - unsigned long op) +int extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end, + struct page *locked_page, + unsigned clear_bits, + unsigned long page_ops) { + struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; int ret; struct page *pages[16]; unsigned long index = start >> PAGE_CACHE_SHIFT; unsigned long end_index = end >> PAGE_CACHE_SHIFT; unsigned long nr_pages = end_index - index + 1; int i; - unsigned long clear_bits = 0; - - if (op & EXTENT_CLEAR_UNLOCK) - clear_bits |= EXTENT_LOCKED; - if (op & EXTENT_CLEAR_DIRTY) - clear_bits |= EXTENT_DIRTY; - - if (op & EXTENT_CLEAR_DELALLOC) - clear_bits |= EXTENT_DELALLOC; clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS); - if (!(op & (EXTENT_CLEAR_UNLOCK_PAGE | EXTENT_CLEAR_DIRTY | - EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK | - EXTENT_SET_PRIVATE2))) + if (page_ops == 0) return 0; + if ((page_ops & PAGE_SET_ERROR) && nr_pages > 0) + mapping_set_error(inode->i_mapping, -EIO); + while (nr_pages > 0) { ret = find_get_pages_contig(inode->i_mapping, index, min_t(unsigned long, nr_pages, ARRAY_SIZE(pages)), pages); for (i = 0; i < ret; i++) { - if (op & EXTENT_SET_PRIVATE2) + if (page_ops & PAGE_SET_PRIVATE2) SetPagePrivate2(pages[i]); if (pages[i] == locked_page) { page_cache_release(pages[i]); continue; } - if (op & EXTENT_CLEAR_DIRTY) + if (page_ops & PAGE_CLEAR_DIRTY) clear_page_dirty_for_io(pages[i]); - if (op & EXTENT_SET_WRITEBACK) + if (page_ops & PAGE_SET_WRITEBACK) set_page_writeback(pages[i]); - if (op & EXTENT_END_WRITEBACK) + if (page_ops & PAGE_SET_ERROR) + SetPageError(pages[i]); + if (page_ops & PAGE_END_WRITEBACK) end_page_writeback(pages[i]); - if (op & EXTENT_CLEAR_UNLOCK_PAGE) + if (page_ops & PAGE_UNLOCK) unlock_page(pages[i]); page_cache_release(pages[i]); } @@ -1722,7 +1879,7 @@ */ u64 count_range_bits(struct extent_io_tree *tree, u64 *start, u64 search_end, u64 max_bytes, - unsigned long bits, int contig) + unsigned bits, int contig) { struct rb_node *node; struct extent_state *state; @@ -1731,10 +1888,8 @@ u64 last = 0; int found = 0; - if (search_end <= cur_start) { - WARN_ON(1); + if (WARN_ON(search_end <= cur_start)) return 0; - } spin_lock(&tree->lock); if (cur_start == 0 && bits == EXTENT_DIRTY) { @@ -1781,7 +1936,7 @@ * set the private field for a given byte offset in the tree. If there isn't * an extent_state there already, this does nothing. */ -int set_state_private(struct extent_io_tree *tree, u64 start, u64 private) +static int set_state_private(struct extent_io_tree *tree, u64 start, u64 private) { struct rb_node *node; struct extent_state *state; @@ -1808,64 +1963,6 @@ return ret; } -void extent_cache_csums_dio(struct extent_io_tree *tree, u64 start, u32 csums[], - int count) -{ - struct rb_node *node; - struct extent_state *state; - - spin_lock(&tree->lock); - /* - * this search will find all the extents that end after - * our range starts. - */ - node = tree_search(tree, start); - BUG_ON(!node); - - state = rb_entry(node, struct extent_state, rb_node); - BUG_ON(state->start != start); - - while (count) { - state->private = *csums++; - count--; - state = next_state(state); - } - spin_unlock(&tree->lock); -} - -static inline u64 __btrfs_get_bio_offset(struct bio *bio, int bio_index) -{ - struct bio_vec *bvec = bio->bi_io_vec + bio_index; - - return page_offset(bvec->bv_page) + bvec->bv_offset; -} - -void extent_cache_csums(struct extent_io_tree *tree, struct bio *bio, int bio_index, - u32 csums[], int count) -{ - struct rb_node *node; - struct extent_state *state = NULL; - u64 start; - - spin_lock(&tree->lock); - do { - start = __btrfs_get_bio_offset(bio, bio_index); - if (state == NULL || state->start != start) { - node = tree_search(tree, start); - BUG_ON(!node); - - state = rb_entry(node, struct extent_state, rb_node); - BUG_ON(state->start != start); - } - state->private = *csums++; - count--; - bio_index++; - - state = next_state(state); - } while (count); - spin_unlock(&tree->lock); -} - int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private) { struct rb_node *node; @@ -1900,14 +1997,14 @@ * range is found set. */ int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end, - unsigned long bits, int filled, struct extent_state *cached) + unsigned bits, int filled, struct extent_state *cached) { struct extent_state *state = NULL; struct rb_node *node; int bitset = 0; spin_lock(&tree->lock); - if (cached && cached->tree && cached->start <= start && + if (cached && extent_state_in_tree(cached) && cached->start <= start && cached->end > start) node = &cached->rb_node; else @@ -1961,27 +2058,7 @@ SetPageUptodate(page); } -/* - * When IO fails, either with EIO or csum verification fails, we - * try other mirrors that might have a good copy of the data. This - * io_failure_record is used to record state as we go through all the - * mirrors. If another mirror has good data, the page is set up to date - * and things continue. If a good mirror can't be found, the original - * bio end_io callback is called to indicate things have failed. - */ -struct io_failure_record { - struct page *page; - u64 start; - u64 len; - u64 logical; - unsigned long bio_flags; - int this_mirror; - int failed_mirror; - int in_validation; -}; - -static int free_io_failure(struct inode *inode, struct io_failure_record *rec, - int did_repair) +int free_io_failure(struct inode *inode, struct io_failure_record *rec) { int ret; int err = 0; @@ -2004,11 +2081,6 @@ return err; } -static void repair_io_failure_callback(struct bio *bio, int err) -{ - complete(bio->bi_private); -} - /* * this bypasses the standard btrfs submit functions deliberately, as * the standard behavior is to write all copies in a raid setup. here we only @@ -2019,19 +2091,19 @@ * currently, there can be no more than two copies of every data bit. thus, * exactly one rewrite is required. */ -int repair_io_failure(struct btrfs_fs_info *fs_info, u64 start, - u64 length, u64 logical, struct page *page, - int mirror_num) +int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical, + struct page *page, unsigned int pg_offset, int mirror_num) { + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; struct bio *bio; struct btrfs_device *dev; - DECLARE_COMPLETION_ONSTACK(compl); u64 map_length = 0; u64 sector; struct btrfs_bio *bbio = NULL; struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree; int ret; + ASSERT(!(fs_info->sb->s_flags & MS_RDONLY)); BUG_ON(!mirror_num); /* we can't repair anything in raid56 yet */ @@ -2041,9 +2113,7 @@ bio = btrfs_io_bio_alloc(GFP_NOFS, 1); if (!bio) return -EIO; - bio->bi_private = &compl; - bio->bi_end_io = repair_io_failure_callback; - bio->bi_size = 0; + bio->bi_iter.bi_size = 0; map_length = length; ret = btrfs_map_block(fs_info, WRITE, logical, @@ -2054,29 +2124,27 @@ } BUG_ON(mirror_num != bbio->mirror_num); sector = bbio->stripes[mirror_num-1].physical >> 9; - bio->bi_sector = sector; + bio->bi_iter.bi_sector = sector; dev = bbio->stripes[mirror_num-1].dev; - kfree(bbio); + btrfs_put_bbio(bbio); if (!dev || !dev->bdev || !dev->writeable) { bio_put(bio); return -EIO; } bio->bi_bdev = dev->bdev; - bio_add_page(bio, page, length, start - page_offset(page)); - btrfsic_submit_bio(WRITE_SYNC, bio); - wait_for_completion(&compl); + bio_add_page(bio, page, length, pg_offset); - if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) { + if (btrfsic_submit_bio_wait(WRITE_SYNC, bio)) { /* try to remap that extent elsewhere? */ bio_put(bio); btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS); return -EIO; } - printk_ratelimited_in_rcu(KERN_INFO "btrfs read error corrected: ino %lu off %llu " - "(dev %s sector %llu)\n", page->mapping->host->i_ino, - start, rcu_str_deref(dev->name), sector); - + btrfs_info_rl_in_rcu(fs_info, + "read error corrected: ino %llu off %llu (dev %s sector %llu)", + btrfs_ino(inode), start, + rcu_str_deref(dev->name), sector); bio_put(bio); return 0; } @@ -2088,10 +2156,15 @@ unsigned long i, num_pages = num_extent_pages(eb->start, eb->len); int ret = 0; + if (root->fs_info->sb->s_flags & MS_RDONLY) + return -EROFS; + for (i = 0; i < num_pages; i++) { - struct page *p = extent_buffer_page(eb, i); - ret = repair_io_failure(root->fs_info, start, PAGE_CACHE_SIZE, - start, p, mirror_num); + struct page *p = eb->pages[i]; + + ret = repair_io_failure(root->fs_info->btree_inode, start, + PAGE_CACHE_SIZE, start, p, + start - page_offset(p), mirror_num); if (ret) break; start += PAGE_CACHE_SIZE; @@ -2104,17 +2177,16 @@ * each time an IO finishes, we do a fast check in the IO failure tree * to see if we need to process or clean up an io_failure_record */ -static int clean_io_failure(u64 start, struct page *page) +int clean_io_failure(struct inode *inode, u64 start, struct page *page, + unsigned int pg_offset) { u64 private; u64 private_failure; struct io_failure_record *failrec; - struct btrfs_fs_info *fs_info; + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; struct extent_state *state; int num_copies; - int did_repair = 0; int ret; - struct inode *inode = page->mapping->host; private = 0; ret = count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private, @@ -2134,9 +2206,10 @@ /* there was no real error, just free the record */ pr_debug("clean_io_failure: freeing dummy error at %llu\n", failrec->start); - did_repair = 1; goto out; } + if (fs_info->sb->s_flags & MS_RDONLY) + goto out; spin_lock(&BTRFS_I(inode)->io_tree.lock); state = find_first_extent_bit_state(&BTRFS_I(inode)->io_tree, @@ -2144,58 +2217,75 @@ EXTENT_LOCKED); spin_unlock(&BTRFS_I(inode)->io_tree.lock); - if (state && state->start == failrec->start) { - fs_info = BTRFS_I(inode)->root->fs_info; + 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) { - ret = repair_io_failure(fs_info, start, failrec->len, - failrec->logical, page, - failrec->failed_mirror); - did_repair = !ret; + repair_io_failure(inode, start, failrec->len, + failrec->logical, page, + pg_offset, failrec->failed_mirror); } - ret = 0; } out: - if (!ret) - ret = free_io_failure(inode, failrec, did_repair); + free_io_failure(inode, failrec); - return ret; + return 0; } /* - * this is a generic handler for readpage errors (default - * readpage_io_failed_hook). if other copies exist, read those and write back - * good data to the failed position. does not investigate in remapping the - * failed extent elsewhere, hoping the device will be smart enough to do this as - * needed + * Can be called when + * - hold extent lock + * - under ordered extent + * - the inode is freeing */ +void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end) +{ + struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree; + struct io_failure_record *failrec; + struct extent_state *state, *next; + + if (RB_EMPTY_ROOT(&failure_tree->state)) + return; + + spin_lock(&failure_tree->lock); + state = find_first_extent_bit_state(failure_tree, start, EXTENT_DIRTY); + while (state) { + if (state->start > end) + break; + + ASSERT(state->end <= end); + + next = next_state(state); + + failrec = (struct io_failure_record *)(unsigned long)state->private; + free_extent_state(state); + kfree(failrec); + + state = next; + } + spin_unlock(&failure_tree->lock); +} -static int bio_readpage_error(struct bio *failed_bio, struct page *page, - u64 start, u64 end, int failed_mirror, - struct extent_state *state) +int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end, + struct io_failure_record **failrec_ret) { - struct io_failure_record *failrec = NULL; + struct io_failure_record *failrec; u64 private; struct extent_map *em; - struct inode *inode = page->mapping->host; struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree; struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; - struct bio *bio; - int num_copies; int ret; - int read_mode; u64 logical; - BUG_ON(failed_bio->bi_rw & REQ_WRITE); - ret = get_state_private(failure_tree, start, &private); if (ret) { failrec = kzalloc(sizeof(*failrec), GFP_NOFS); if (!failrec) return -ENOMEM; + failrec->start = start; failrec->len = end - start + 1; failrec->this_mirror = 0; @@ -2210,16 +2300,16 @@ return -EIO; } - if (em->start > start || em->start + em->len < start) { + if (em->start > start || em->start + em->len <= start) { free_extent_map(em); em = NULL; } read_unlock(&em_tree->lock); - if (!em) { kfree(failrec); return -EIO; } + logical = start - em->start; logical = em->block_start + logical; if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) { @@ -2228,8 +2318,10 @@ extent_set_compress_type(&failrec->bio_flags, em->compress_type); } - pr_debug("bio_readpage_error: (new) logical=%llu, start=%llu, " - "len=%llu\n", logical, start, failrec->len); + + pr_debug("Get IO Failure Record: (new) logical=%llu, start=%llu, len=%llu\n", + logical, start, failrec->len); + failrec->logical = logical; free_extent_map(em); @@ -2249,8 +2341,7 @@ } } else { failrec = (struct io_failure_record *)(unsigned long)private; - pr_debug("bio_readpage_error: (found) logical=%llu, " - "start=%llu, len=%llu, validation=%d\n", + pr_debug("Get IO Failure Record: (found) logical=%llu, start=%llu, len=%llu, validation=%d\n", failrec->logical, failrec->start, failrec->len, failrec->in_validation); /* @@ -2259,6 +2350,17 @@ * clean_io_failure() clean all those errors at once. */ } + + *failrec_ret = failrec; + + return 0; +} + +int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio, + struct io_failure_record *failrec, int failed_mirror) +{ + int num_copies; + num_copies = btrfs_num_copies(BTRFS_I(inode)->root->fs_info, failrec->logical, failrec->len); if (num_copies == 1) { @@ -2267,21 +2369,9 @@ * all the retry and error correction code that follows. no * matter what the error is, it is very likely to persist. */ - pr_debug("bio_readpage_error: cannot repair, num_copies == 1. " - "state=%p, num_copies=%d, next_mirror %d, " - "failed_mirror %d\n", state, num_copies, - failrec->this_mirror, failed_mirror); - free_io_failure(inode, failrec, 0); - return -EIO; - } - - if (!state) { - spin_lock(&tree->lock); - state = find_first_extent_bit_state(tree, failrec->start, - EXTENT_LOCKED); - if (state && state->start != failrec->start) - state = NULL; - spin_unlock(&tree->lock); + pr_debug("Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d\n", + num_copies, failrec->this_mirror, failed_mirror); + return 0; } /* @@ -2301,7 +2391,6 @@ BUG_ON(failrec->in_validation); failrec->in_validation = 1; failrec->this_mirror = failed_mirror; - read_mode = READ_SYNC | REQ_FAILFAST_DEV; } else { /* * we're ready to fulfill a) and b) alongside. get a good copy @@ -2317,37 +2406,111 @@ failrec->this_mirror++; if (failrec->this_mirror == failed_mirror) failrec->this_mirror++; - read_mode = READ_SYNC; } - if (!state || failrec->this_mirror > num_copies) { - pr_debug("bio_readpage_error: (fail) state=%p, num_copies=%d, " - "next_mirror %d, failed_mirror %d\n", state, + if (failrec->this_mirror > num_copies) { + pr_debug("Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d\n", num_copies, failrec->this_mirror, failed_mirror); - free_io_failure(inode, failrec, 0); - return -EIO; + return 0; } + return 1; +} + + +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) +{ + struct bio *bio; + struct btrfs_io_bio *btrfs_failed_bio; + struct btrfs_io_bio *btrfs_bio; + bio = btrfs_io_bio_alloc(GFP_NOFS, 1); - if (!bio) { - free_io_failure(inode, failrec, 0); + if (!bio) + return NULL; + + bio->bi_end_io = endio_func; + bio->bi_iter.bi_sector = failrec->logical >> 9; + bio->bi_bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev; + bio->bi_iter.bi_size = 0; + bio->bi_private = data; + + btrfs_failed_bio = btrfs_io_bio(failed_bio); + if (btrfs_failed_bio->csum) { + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + u16 csum_size = btrfs_super_csum_size(fs_info->super_copy); + + btrfs_bio = btrfs_io_bio(bio); + btrfs_bio->csum = btrfs_bio->csum_inline; + icsum *= csum_size; + memcpy(btrfs_bio->csum, btrfs_failed_bio->csum + icsum, + csum_size); + } + + bio_add_page(bio, page, failrec->len, pg_offset); + + return bio; +} + +/* + * this is a generic handler for readpage errors (default + * readpage_io_failed_hook). if other copies exist, read those and write back + * good data to the failed position. does not investigate in remapping the + * failed extent elsewhere, hoping the device will be smart enough to do this as + * needed + */ + +static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset, + struct page *page, u64 start, u64 end, + int failed_mirror) +{ + struct io_failure_record *failrec; + struct inode *inode = page->mapping->host; + struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; + struct bio *bio; + int read_mode; + int ret; + + BUG_ON(failed_bio->bi_rw & REQ_WRITE); + + ret = btrfs_get_io_failure_record(inode, start, end, &failrec); + if (ret) + return ret; + + ret = btrfs_check_repairable(inode, failed_bio, failrec, failed_mirror); + if (!ret) { + free_io_failure(inode, failrec); return -EIO; } - bio->bi_private = state; - bio->bi_end_io = failed_bio->bi_end_io; - bio->bi_sector = failrec->logical >> 9; - bio->bi_bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev; - bio->bi_size = 0; - bio_add_page(bio, page, failrec->len, start - page_offset(page)); + if (failed_bio->bi_vcnt > 1) + read_mode = READ_SYNC | REQ_FAILFAST_DEV; + else + read_mode = READ_SYNC; + + phy_offset >>= inode->i_sb->s_blocksize_bits; + bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page, + start - page_offset(page), + (int)phy_offset, failed_bio->bi_end_io, + NULL); + if (!bio) { + free_io_failure(inode, failrec); + return -EIO; + } - pr_debug("bio_readpage_error: submitting new read[%#x] to " - "this_mirror=%d, num_copies=%d, in_validation=%d\n", read_mode, - failrec->this_mirror, num_copies, failrec->in_validation); + pr_debug("Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d\n", + read_mode, failrec->this_mirror, failrec->in_validation); ret = tree->ops->submit_bio_hook(inode, read_mode, bio, failrec->this_mirror, failrec->bio_flags, 0); + if (ret) { + free_io_failure(inode, failrec); + bio_put(bio); + } + return ret; } @@ -2386,43 +2549,57 @@ * Scheduling is not allowed, so the extent state tree is expected * to have one and only one object corresponding to this IO. */ -static void end_bio_extent_writepage(struct bio *bio, int err) +static void end_bio_extent_writepage(struct bio *bio) { - struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; - struct extent_io_tree *tree; + struct bio_vec *bvec; u64 start; u64 end; + int i; - do { + bio_for_each_segment_all(bvec, bio, i) { struct page *page = bvec->bv_page; - tree = &BTRFS_I(page->mapping->host)->io_tree; /* We always issue full-page reads, but if some block * in a page fails to read, blk_update_request() will * advance bv_offset and adjust bv_len to compensate. * Print a warning for nonzero offsets, and an error * if they don't add up to a full page. */ - if (bvec->bv_offset || bvec->bv_len != PAGE_CACHE_SIZE) - printk("%s page write in btrfs with offset %u and length %u\n", - bvec->bv_offset + bvec->bv_len != PAGE_CACHE_SIZE - ? KERN_ERR "partial" : KERN_INFO "incomplete", - bvec->bv_offset, bvec->bv_len); + if (bvec->bv_offset || bvec->bv_len != PAGE_CACHE_SIZE) { + if (bvec->bv_offset + bvec->bv_len != PAGE_CACHE_SIZE) + btrfs_err(BTRFS_I(page->mapping->host)->root->fs_info, + "partial page write in btrfs with offset %u and length %u", + bvec->bv_offset, bvec->bv_len); + else + btrfs_info(BTRFS_I(page->mapping->host)->root->fs_info, + "incomplete page write in btrfs with offset %u and " + "length %u", + bvec->bv_offset, bvec->bv_len); + } start = page_offset(page); end = start + bvec->bv_offset + bvec->bv_len - 1; - if (--bvec >= bio->bi_io_vec) - prefetchw(&bvec->bv_page->flags); - - if (end_extent_writepage(page, err, start, end)) + if (end_extent_writepage(page, bio->bi_error, start, end)) continue; end_page_writeback(page); - } while (bvec >= bio->bi_io_vec); + } bio_put(bio); } +static void +endio_readpage_release_extent(struct extent_io_tree *tree, u64 start, u64 len, + int uptodate) +{ + struct extent_state *cached = NULL; + u64 end = start + len - 1; + + if (uptodate && tree->track_uptodate) + set_extent_uptodate(tree, start, end, &cached, GFP_ATOMIC); + unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC); +} + /* * after a readpage IO is done, we need to: * clear the uptodate bits on error @@ -2434,75 +2611,72 @@ * Scheduling is not allowed, so the extent state tree is expected * to have one and only one object corresponding to this IO. */ -static void end_bio_extent_readpage(struct bio *bio, int err) +static void end_bio_extent_readpage(struct bio *bio) { - int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1; - struct bio_vec *bvec = bio->bi_io_vec; + struct bio_vec *bvec; + int uptodate = !bio->bi_error; + struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); struct extent_io_tree *tree; + u64 offset = 0; u64 start; u64 end; + u64 len; + u64 extent_start = 0; + u64 extent_len = 0; int mirror; int ret; + int i; - if (err) - uptodate = 0; - - do { + bio_for_each_segment_all(bvec, bio, i) { struct page *page = bvec->bv_page; - struct extent_state *cached = NULL; - struct extent_state *state; - struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); + struct inode *inode = page->mapping->host; pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, " - "mirror=%lu\n", (u64)bio->bi_sector, err, - io_bio->mirror_num); - tree = &BTRFS_I(page->mapping->host)->io_tree; + "mirror=%u\n", (u64)bio->bi_iter.bi_sector, + bio->bi_error, io_bio->mirror_num); + tree = &BTRFS_I(inode)->io_tree; /* We always issue full-page reads, but if some block * in a page fails to read, blk_update_request() will * advance bv_offset and adjust bv_len to compensate. * Print a warning for nonzero offsets, and an error * if they don't add up to a full page. */ - if (bvec->bv_offset || bvec->bv_len != PAGE_CACHE_SIZE) - printk("%s page read in btrfs with offset %u and length %u\n", - bvec->bv_offset + bvec->bv_len != PAGE_CACHE_SIZE - ? KERN_ERR "partial" : KERN_INFO "incomplete", - bvec->bv_offset, bvec->bv_len); + if (bvec->bv_offset || bvec->bv_len != PAGE_CACHE_SIZE) { + if (bvec->bv_offset + bvec->bv_len != PAGE_CACHE_SIZE) + btrfs_err(BTRFS_I(page->mapping->host)->root->fs_info, + "partial page read in btrfs with offset %u and length %u", + bvec->bv_offset, bvec->bv_len); + else + btrfs_info(BTRFS_I(page->mapping->host)->root->fs_info, + "incomplete page read in btrfs with offset %u and " + "length %u", + bvec->bv_offset, bvec->bv_len); + } start = page_offset(page); end = start + bvec->bv_offset + bvec->bv_len - 1; - - if (++bvec <= bvec_end) - prefetchw(&bvec->bv_page->flags); - - spin_lock(&tree->lock); - state = find_first_extent_bit_state(tree, start, EXTENT_LOCKED); - if (state && state->start == start) { - /* - * take a reference on the state, unlock will drop - * the ref - */ - cache_state(state, &cached); - } - spin_unlock(&tree->lock); + len = bvec->bv_len; mirror = io_bio->mirror_num; - if (uptodate && tree->ops && tree->ops->readpage_end_io_hook) { - ret = tree->ops->readpage_end_io_hook(page, start, end, - state, mirror); + if (likely(uptodate && tree->ops && + tree->ops->readpage_end_io_hook)) { + ret = tree->ops->readpage_end_io_hook(io_bio, offset, + page, start, end, + mirror); if (ret) uptodate = 0; else - clean_io_failure(start, page); + clean_io_failure(inode, start, page, 0); } - if (!uptodate && tree->ops && tree->ops->readpage_io_failed_hook) { + if (likely(uptodate)) + goto readpage_ok; + + if (tree->ops && tree->ops->readpage_io_failed_hook) { ret = tree->ops->readpage_io_failed_hook(page, mirror); - if (!ret && !err && - test_bit(BIO_UPTODATE, &bio->bi_flags)) + if (!ret && !bio->bi_error) uptodate = 1; - } else if (!uptodate) { + } else { /* * The generic bio_readpage_error handles errors the * following way: If possible, new read requests are @@ -2513,32 +2687,60 @@ * can't handle the error it will return -EIO and we * remain responsible for that page. */ - ret = bio_readpage_error(bio, page, start, end, mirror, NULL); + ret = bio_readpage_error(bio, offset, page, start, end, + mirror); if (ret == 0) { - uptodate = - test_bit(BIO_UPTODATE, &bio->bi_flags); - if (err) - uptodate = 0; - uncache_state(&cached); + uptodate = !bio->bi_error; + offset += len; continue; } } - - if (uptodate && tree->track_uptodate) { - set_extent_uptodate(tree, start, end, &cached, - GFP_ATOMIC); - } - unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC); - - if (uptodate) { +readpage_ok: + if (likely(uptodate)) { + loff_t i_size = i_size_read(inode); + pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned off; + + /* Zero out the end if this page straddles i_size */ + off = i_size & (PAGE_CACHE_SIZE-1); + if (page->index == end_index && off) + zero_user_segment(page, off, PAGE_CACHE_SIZE); SetPageUptodate(page); } else { ClearPageUptodate(page); SetPageError(page); } unlock_page(page); - } while (bvec <= bvec_end); + offset += len; + if (unlikely(!uptodate)) { + if (extent_len) { + endio_readpage_release_extent(tree, + extent_start, + extent_len, 1); + extent_start = 0; + extent_len = 0; + } + endio_readpage_release_extent(tree, start, + end - start + 1, 0); + } else if (!extent_len) { + extent_start = start; + extent_len = end + 1 - start; + } else if (extent_start + extent_len == start) { + extent_len += end + 1 - start; + } else { + endio_readpage_release_extent(tree, extent_start, + extent_len, uptodate); + extent_start = start; + extent_len = end + 1 - start; + } + } + + if (extent_len) + endio_readpage_release_extent(tree, extent_start, extent_len, + uptodate); + if (io_bio->end_io) + io_bio->end_io(io_bio, bio->bi_error); bio_put(bio); } @@ -2550,6 +2752,7 @@ btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, gfp_t gfp_flags) { + struct btrfs_io_bio *btrfs_bio; struct bio *bio; bio = bio_alloc_bioset(gfp_flags, nr_vecs, btrfs_bioset); @@ -2562,23 +2765,45 @@ } if (bio) { - bio->bi_size = 0; bio->bi_bdev = bdev; - bio->bi_sector = first_sector; + bio->bi_iter.bi_sector = first_sector; + btrfs_bio = btrfs_io_bio(bio); + btrfs_bio->csum = NULL; + btrfs_bio->csum_allocated = NULL; + btrfs_bio->end_io = NULL; } return bio; } struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask) { - return bio_clone_bioset(bio, gfp_mask, btrfs_bioset); -} + struct btrfs_io_bio *btrfs_bio; + struct bio *new; + new = bio_clone_bioset(bio, gfp_mask, btrfs_bioset); + if (new) { + btrfs_bio = btrfs_io_bio(new); + btrfs_bio->csum = NULL; + btrfs_bio->csum_allocated = NULL; + btrfs_bio->end_io = NULL; + } + return new; +} /* this also allocates from the btrfs_bioset */ struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) { - return bio_alloc_bioset(gfp_mask, nr_iovecs, btrfs_bioset); + struct btrfs_io_bio *btrfs_bio; + struct bio *bio; + + bio = bio_alloc_bioset(gfp_mask, nr_iovecs, btrfs_bioset); + if (bio) { + btrfs_bio = btrfs_io_bio(bio); + btrfs_bio->csum = NULL; + btrfs_bio->csum_allocated = NULL; + btrfs_bio->end_io = NULL; + } + return bio; } @@ -2603,8 +2828,6 @@ else btrfsic_submit_bio(rw, bio); - if (bio_flagged(bio, BIO_EOPNOTSUPP)) - ret = -EOPNOTSUPP; bio_put(bio); return ret; } @@ -2623,6 +2846,7 @@ } static int submit_extent_page(int rw, struct extent_io_tree *tree, + struct writeback_control *wbc, struct page *page, sector_t sector, size_t size, unsigned long offset, struct block_device *bdev, @@ -2631,47 +2855,52 @@ bio_end_io_t end_io_func, int mirror_num, unsigned long prev_bio_flags, - unsigned long bio_flags) + unsigned long bio_flags, + bool force_bio_submit) { int ret = 0; struct bio *bio; - int nr; int contig = 0; - int this_compressed = bio_flags & EXTENT_BIO_COMPRESSED; int old_compressed = prev_bio_flags & EXTENT_BIO_COMPRESSED; size_t page_size = min_t(size_t, size, PAGE_CACHE_SIZE); if (bio_ret && *bio_ret) { bio = *bio_ret; if (old_compressed) - contig = bio->bi_sector == sector; + contig = bio->bi_iter.bi_sector == sector; else contig = bio_end_sector(bio) == sector; if (prev_bio_flags != bio_flags || !contig || + force_bio_submit || merge_bio(rw, tree, page, offset, page_size, bio, bio_flags) || bio_add_page(bio, page, page_size, offset) < page_size) { ret = submit_one_bio(rw, bio, mirror_num, prev_bio_flags); - if (ret < 0) + if (ret < 0) { + *bio_ret = NULL; return ret; + } bio = NULL; } else { + if (wbc) + wbc_account_io(wbc, page, page_size); return 0; } } - if (this_compressed) - nr = BIO_MAX_PAGES; - else - nr = bio_get_nr_vecs(bdev); - bio = btrfs_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH); + bio = btrfs_bio_alloc(bdev, sector, BIO_MAX_PAGES, + GFP_NOFS | __GFP_HIGH); if (!bio) return -ENOMEM; bio_add_page(bio, page, page_size, offset); bio->bi_end_io = end_io_func; bio->bi_private = tree; + if (wbc) { + wbc_init_bio(wbc, bio); + wbc_account_io(wbc, page, page_size); + } if (bio_ret) *bio_ret = bio; @@ -2702,17 +2931,46 @@ } } +static struct extent_map * +__get_extent_map(struct inode *inode, struct page *page, size_t pg_offset, + u64 start, u64 len, get_extent_t *get_extent, + struct extent_map **em_cached) +{ + struct extent_map *em; + + if (em_cached && *em_cached) { + em = *em_cached; + if (extent_map_in_tree(em) && start >= em->start && + start < extent_map_end(em)) { + atomic_inc(&em->refs); + return em; + } + + free_extent_map(em); + *em_cached = NULL; + } + + em = get_extent(inode, page, pg_offset, start, len, 0); + if (em_cached && !IS_ERR_OR_NULL(em)) { + BUG_ON(*em_cached); + atomic_inc(&em->refs); + *em_cached = em; + } + return em; +} /* * basic readpage implementation. Locked extent state structs are inserted * into the tree that are removed when the IO is done (by the end_io * handlers) * XXX JDM: This needs looking at to ensure proper page locking */ -static int __extent_read_full_page(struct extent_io_tree *tree, - struct page *page, - get_extent_t *get_extent, - struct bio **bio, int mirror_num, - unsigned long *bio_flags, int rw) +static int __do_readpage(struct extent_io_tree *tree, + struct page *page, + get_extent_t *get_extent, + struct extent_map **em_cached, + struct bio **bio, int mirror_num, + unsigned long *bio_flags, int rw, + u64 *prev_em_start) { struct inode *inode = page->mapping->host; u64 start = page_offset(page); @@ -2726,35 +2984,26 @@ sector_t sector; struct extent_map *em; struct block_device *bdev; - struct btrfs_ordered_extent *ordered; int ret; int nr = 0; + int parent_locked = *bio_flags & EXTENT_BIO_PARENT_LOCKED; size_t pg_offset = 0; size_t iosize; size_t disk_io_size; size_t blocksize = inode->i_sb->s_blocksize; - unsigned long this_bio_flag = 0; + unsigned long this_bio_flag = *bio_flags & EXTENT_BIO_PARENT_LOCKED; set_page_extent_mapped(page); + end = page_end; if (!PageUptodate(page)) { if (cleancache_get_page(page) == 0) { BUG_ON(blocksize != PAGE_SIZE); + unlock_extent(tree, start, end); goto out; } } - end = page_end; - while (1) { - lock_extent(tree, start, end); - ordered = btrfs_lookup_ordered_extent(inode, start); - if (!ordered) - break; - unlock_extent(tree, start, end); - btrfs_start_ordered_extent(inode, ordered, 1); - btrfs_put_ordered_extent(ordered); - } - if (page->index == last_byte >> PAGE_CACHE_SHIFT) { char *userpage; size_t zero_offset = last_byte & (PAGE_CACHE_SIZE - 1); @@ -2769,6 +3018,7 @@ } while (cur <= end) { unsigned long pnr = (last_byte >> PAGE_CACHE_SHIFT) + 1; + bool force_bio_submit = false; if (cur >= last_byte) { char *userpage; @@ -2781,15 +3031,18 @@ kunmap_atomic(userpage); set_extent_uptodate(tree, cur, cur + iosize - 1, &cached, GFP_NOFS); - unlock_extent_cached(tree, cur, cur + iosize - 1, - &cached, GFP_NOFS); + if (!parent_locked) + unlock_extent_cached(tree, cur, + cur + iosize - 1, + &cached, GFP_NOFS); break; } - em = get_extent(inode, page, pg_offset, cur, - end - cur + 1, 0); + em = __get_extent_map(inode, page, pg_offset, cur, + end - cur + 1, get_extent, em_cached); if (IS_ERR_OR_NULL(em)) { SetPageError(page); - unlock_extent(tree, cur, end); + if (!parent_locked) + unlock_extent(tree, cur, end); break; } extent_offset = cur - em->start; @@ -2797,7 +3050,7 @@ BUG_ON(end < cur); if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) { - this_bio_flag = EXTENT_BIO_COMPRESSED; + this_bio_flag |= EXTENT_BIO_COMPRESSED; extent_set_compress_type(&this_bio_flag, em->compress_type); } @@ -2816,6 +3069,49 @@ block_start = em->block_start; if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) block_start = EXTENT_MAP_HOLE; + + /* + * If we have a file range that points to a compressed extent + * and it's followed by a consecutive file range that points to + * to the same compressed extent (possibly with a different + * offset and/or length, so it either points to the whole extent + * or only part of it), we must make sure we do not submit a + * single bio to populate the pages for the 2 ranges because + * this makes the compressed extent read zero out the pages + * belonging to the 2nd range. Imagine the following scenario: + * + * File layout + * [0 - 8K] [8K - 24K] + * | | + * | | + * points to extent X, points to extent X, + * offset 4K, length of 8K offset 0, length 16K + * + * [extent X, compressed length = 4K uncompressed length = 16K] + * + * If the bio to read the compressed extent covers both ranges, + * it will decompress extent X into the pages belonging to the + * first range and then it will stop, zeroing out the remaining + * pages that belong to the other range that points to extent X. + * So here we make sure we submit 2 bios, one for the first + * range and another one for the third range. Both will target + * the same physical extent from disk, but we can't currently + * make the compressed bio endio callback populate the pages + * for both ranges because each compressed bio is tightly + * coupled with a single extent map, and each range can have + * an extent map with a different offset value relative to the + * uncompressed data of our extent and different lengths. This + * is a corner case so we prioritize correctness over + * non-optimal behavior (submitting 2 bios for the same extent). + */ + if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags) && + prev_em_start && *prev_em_start != (u64)-1 && + *prev_em_start != em->orig_start) + force_bio_submit = true; + + if (prev_em_start) + *prev_em_start = em->orig_start; + free_extent_map(em); em = NULL; @@ -2831,8 +3127,12 @@ set_extent_uptodate(tree, cur, cur + iosize - 1, &cached, GFP_NOFS); - unlock_extent_cached(tree, cur, cur + iosize - 1, - &cached, GFP_NOFS); + if (parent_locked) + free_extent_state(cached); + else + unlock_extent_cached(tree, cur, + cur + iosize - 1, + &cached, GFP_NOFS); cur = cur + iosize; pg_offset += iosize; continue; @@ -2841,7 +3141,8 @@ if (test_range_bit(tree, cur, cur_end, EXTENT_UPTODATE, 1, NULL)) { check_page_uptodate(tree, page); - unlock_extent(tree, cur, cur + iosize - 1); + if (!parent_locked) + unlock_extent(tree, cur, cur + iosize - 1); cur = cur + iosize; pg_offset += iosize; continue; @@ -2851,25 +3152,28 @@ */ if (block_start == EXTENT_MAP_INLINE) { SetPageError(page); - unlock_extent(tree, cur, cur + iosize - 1); + if (!parent_locked) + unlock_extent(tree, cur, cur + iosize - 1); cur = cur + iosize; pg_offset += iosize; continue; } pnr -= page->index; - ret = submit_extent_page(rw, tree, page, + ret = submit_extent_page(rw, tree, NULL, page, sector, disk_io_size, pg_offset, bdev, bio, pnr, end_bio_extent_readpage, mirror_num, *bio_flags, - this_bio_flag); + this_bio_flag, + force_bio_submit); if (!ret) { nr++; *bio_flags = this_bio_flag; } else { SetPageError(page); - unlock_extent(tree, cur, cur + iosize - 1); + if (!parent_locked) + unlock_extent(tree, cur, cur + iosize - 1); } cur = cur + iosize; pg_offset += iosize; @@ -2883,6 +3187,107 @@ return 0; } +static inline void __do_contiguous_readpages(struct extent_io_tree *tree, + struct page *pages[], int nr_pages, + u64 start, u64 end, + get_extent_t *get_extent, + struct extent_map **em_cached, + struct bio **bio, int mirror_num, + unsigned long *bio_flags, int rw, + u64 *prev_em_start) +{ + struct inode *inode; + struct btrfs_ordered_extent *ordered; + int index; + + inode = pages[0]->mapping->host; + while (1) { + lock_extent(tree, start, end); + ordered = btrfs_lookup_ordered_range(inode, start, + end - start + 1); + if (!ordered) + break; + unlock_extent(tree, start, end); + btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); + } + + for (index = 0; index < nr_pages; index++) { + __do_readpage(tree, pages[index], get_extent, em_cached, bio, + mirror_num, bio_flags, rw, prev_em_start); + page_cache_release(pages[index]); + } +} + +static void __extent_readpages(struct extent_io_tree *tree, + struct page *pages[], + int nr_pages, get_extent_t *get_extent, + struct extent_map **em_cached, + struct bio **bio, int mirror_num, + unsigned long *bio_flags, int rw, + u64 *prev_em_start) +{ + u64 start = 0; + u64 end = 0; + u64 page_start; + int index; + int first_index = 0; + + for (index = 0; index < nr_pages; index++) { + page_start = page_offset(pages[index]); + if (!end) { + start = page_start; + end = start + PAGE_CACHE_SIZE - 1; + first_index = index; + } else if (end + 1 == page_start) { + end += PAGE_CACHE_SIZE; + } else { + __do_contiguous_readpages(tree, &pages[first_index], + index - first_index, start, + end, get_extent, em_cached, + bio, mirror_num, bio_flags, + rw, prev_em_start); + start = page_start; + end = start + PAGE_CACHE_SIZE - 1; + first_index = index; + } + } + + if (end) + __do_contiguous_readpages(tree, &pages[first_index], + index - first_index, start, + end, get_extent, em_cached, bio, + mirror_num, bio_flags, rw, + prev_em_start); +} + +static int __extent_read_full_page(struct extent_io_tree *tree, + struct page *page, + get_extent_t *get_extent, + struct bio **bio, int mirror_num, + unsigned long *bio_flags, int rw) +{ + struct inode *inode = page->mapping->host; + struct btrfs_ordered_extent *ordered; + u64 start = page_offset(page); + u64 end = start + PAGE_CACHE_SIZE - 1; + int ret; + + while (1) { + lock_extent(tree, start, end); + ordered = btrfs_lookup_ordered_extent(inode, start); + if (!ordered) + break; + unlock_extent(tree, start, end); + btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); + } + + ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num, + bio_flags, rw, NULL); + return ret; +} + int extent_read_full_page(struct extent_io_tree *tree, struct page *page, get_extent_t *get_extent, int mirror_num) { @@ -2897,6 +3302,20 @@ return ret; } +int extent_read_full_page_nolock(struct extent_io_tree *tree, struct page *page, + get_extent_t *get_extent, int mirror_num) +{ + struct bio *bio = NULL; + unsigned long bio_flags = EXTENT_BIO_PARENT_LOCKED; + int ret; + + ret = __do_readpage(tree, page, get_extent, NULL, &bio, mirror_num, + &bio_flags, READ, NULL); + if (bio) + ret = submit_one_bio(READ, bio, mirror_num, bio_flags); + return ret; +} + static noinline void update_nr_written(struct page *page, struct writeback_control *wbc, unsigned long nr_written) @@ -2908,143 +3327,130 @@ } /* - * the writepage semantics are similar to regular writepage. extent - * records are inserted to lock ranges in the tree, and as dirty areas - * are found, they are marked writeback. Then the lock bits are removed - * and the end_io handler clears the writeback ranges + * helper for __extent_writepage, doing all of the delayed allocation setup. + * + * This returns 1 if our fill_delalloc function did all the work required + * to write the page (copy into inline extent). In this case the IO has + * been started and the page is already unlocked. + * + * This returns 0 if all went well (page still locked) + * This returns < 0 if there were errors (page still locked) */ -static int __extent_writepage(struct page *page, struct writeback_control *wbc, - void *data) +static noinline_for_stack int writepage_delalloc(struct inode *inode, + struct page *page, struct writeback_control *wbc, + struct extent_page_data *epd, + u64 delalloc_start, + unsigned long *nr_written) +{ + struct extent_io_tree *tree = epd->tree; + u64 page_end = delalloc_start + PAGE_CACHE_SIZE - 1; + u64 nr_delalloc; + u64 delalloc_to_write = 0; + u64 delalloc_end = 0; + int ret; + int page_started = 0; + + if (epd->extent_locked || !tree->ops || !tree->ops->fill_delalloc) + return 0; + + while (delalloc_end < page_end) { + nr_delalloc = find_lock_delalloc_range(inode, tree, + page, + &delalloc_start, + &delalloc_end, + BTRFS_MAX_EXTENT_SIZE); + if (nr_delalloc == 0) { + delalloc_start = delalloc_end + 1; + continue; + } + ret = tree->ops->fill_delalloc(inode, page, + delalloc_start, + delalloc_end, + &page_started, + nr_written); + /* File system has been set read-only */ + if (ret) { + SetPageError(page); + /* fill_delalloc should be return < 0 for error + * but just in case, we use > 0 here meaning the + * IO is started, so we don't want to return > 0 + * unless things are going well. + */ + ret = ret < 0 ? ret : -EIO; + goto done; + } + /* + * delalloc_end is already one less than the total + * length, so we don't subtract one from + * PAGE_CACHE_SIZE + */ + delalloc_to_write += (delalloc_end - delalloc_start + + PAGE_CACHE_SIZE) >> + PAGE_CACHE_SHIFT; + delalloc_start = delalloc_end + 1; + } + if (wbc->nr_to_write < delalloc_to_write) { + int thresh = 8192; + + if (delalloc_to_write < thresh * 2) + thresh = delalloc_to_write; + wbc->nr_to_write = min_t(u64, delalloc_to_write, + thresh); + } + + /* did the fill delalloc function already unlock and start + * the IO? + */ + if (page_started) { + /* + * we've unlocked the page, so we can't update + * the mapping's writeback index, just update + * nr_to_write. + */ + wbc->nr_to_write -= *nr_written; + return 1; + } + + ret = 0; + +done: + return ret; +} + +/* + * helper for __extent_writepage. This calls the writepage start hooks, + * and does the loop to map the page into extents and bios. + * + * We return 1 if the IO is started and the page is unlocked, + * 0 if all went well (page still locked) + * < 0 if there were errors (page still locked) + */ +static noinline_for_stack int __extent_writepage_io(struct inode *inode, + struct page *page, + struct writeback_control *wbc, + struct extent_page_data *epd, + loff_t i_size, + unsigned long nr_written, + int write_flags, int *nr_ret) { - struct inode *inode = page->mapping->host; - struct extent_page_data *epd = data; struct extent_io_tree *tree = epd->tree; u64 start = page_offset(page); - u64 delalloc_start; u64 page_end = start + PAGE_CACHE_SIZE - 1; u64 end; u64 cur = start; u64 extent_offset; - u64 last_byte = i_size_read(inode); u64 block_start; u64 iosize; sector_t sector; struct extent_state *cached_state = NULL; struct extent_map *em; struct block_device *bdev; - int ret; - int nr = 0; size_t pg_offset = 0; size_t blocksize; - loff_t i_size = i_size_read(inode); - unsigned long end_index = i_size >> PAGE_CACHE_SHIFT; - u64 nr_delalloc; - u64 delalloc_end; - int page_started; - int compressed; - int write_flags; - unsigned long nr_written = 0; - bool fill_delalloc = true; - - if (wbc->sync_mode == WB_SYNC_ALL) - write_flags = WRITE_SYNC; - else - write_flags = WRITE; - - trace___extent_writepage(page, inode, wbc); - - WARN_ON(!PageLocked(page)); - - ClearPageError(page); - - pg_offset = i_size & (PAGE_CACHE_SIZE - 1); - if (page->index > end_index || - (page->index == end_index && !pg_offset)) { - page->mapping->a_ops->invalidatepage(page, 0); - unlock_page(page); - return 0; - } - - if (page->index == end_index) { - char *userpage; - - userpage = kmap_atomic(page); - memset(userpage + pg_offset, 0, - PAGE_CACHE_SIZE - pg_offset); - kunmap_atomic(userpage); - flush_dcache_page(page); - } - pg_offset = 0; - - set_page_extent_mapped(page); - - if (!tree->ops || !tree->ops->fill_delalloc) - fill_delalloc = false; - - delalloc_start = start; - delalloc_end = 0; - page_started = 0; - if (!epd->extent_locked && fill_delalloc) { - u64 delalloc_to_write = 0; - /* - * make sure the wbc mapping index is at least updated - * to this page. - */ - update_nr_written(page, wbc, 0); - - while (delalloc_end < page_end) { - nr_delalloc = find_lock_delalloc_range(inode, tree, - page, - &delalloc_start, - &delalloc_end, - 128 * 1024 * 1024); - if (nr_delalloc == 0) { - delalloc_start = delalloc_end + 1; - continue; - } - ret = tree->ops->fill_delalloc(inode, page, - delalloc_start, - delalloc_end, - &page_started, - &nr_written); - /* File system has been set read-only */ - if (ret) { - SetPageError(page); - goto done; - } - /* - * delalloc_end is already one less than the total - * length, so we don't subtract one from - * PAGE_CACHE_SIZE - */ - delalloc_to_write += (delalloc_end - delalloc_start + - PAGE_CACHE_SIZE) >> - PAGE_CACHE_SHIFT; - delalloc_start = delalloc_end + 1; - } - if (wbc->nr_to_write < delalloc_to_write) { - int thresh = 8192; - - if (delalloc_to_write < thresh * 2) - thresh = delalloc_to_write; - wbc->nr_to_write = min_t(u64, delalloc_to_write, - thresh); - } + int ret = 0; + int nr = 0; + bool compressed; - /* did the fill delalloc function already unlock and start - * the IO? - */ - if (page_started) { - ret = 0; - /* - * we've unlocked the page, so we can't update - * the mapping's writeback index, just update - * nr_to_write. - */ - wbc->nr_to_write -= nr_written; - goto done_unlocked; - } - } if (tree->ops && tree->ops->writepage_start_hook) { ret = tree->ops->writepage_start_hook(page, start, page_end); @@ -3054,9 +3460,10 @@ wbc->pages_skipped++; else redirty_page_for_writepage(wbc, page); + update_nr_written(page, wbc, nr_written); unlock_page(page); - ret = 0; + ret = 1; goto done_unlocked; } } @@ -3068,7 +3475,7 @@ update_nr_written(page, wbc, nr_written + 1); end = page_end; - if (last_byte <= start) { + if (i_size <= start) { if (tree->ops && tree->ops->writepage_end_io_hook) tree->ops->writepage_end_io_hook(page, start, page_end, NULL, 1); @@ -3078,7 +3485,8 @@ blocksize = inode->i_sb->s_blocksize; while (cur <= end) { - if (cur >= last_byte) { + u64 em_end; + if (cur >= i_size) { if (tree->ops && tree->ops->writepage_end_io_hook) tree->ops->writepage_end_io_hook(page, cur, page_end, NULL, 1); @@ -3088,13 +3496,15 @@ end - cur + 1, 1); if (IS_ERR_OR_NULL(em)) { SetPageError(page); + ret = PTR_ERR_OR_ZERO(em); break; } extent_offset = cur - em->start; - BUG_ON(extent_map_end(em) <= cur); + em_end = extent_map_end(em); + BUG_ON(em_end <= cur); BUG_ON(end < cur); - iosize = min(extent_map_end(em) - cur, end - cur + 1); + iosize = min(em_end - cur, end - cur + 1); iosize = ALIGN(iosize, blocksize); sector = (em->block_start + extent_offset) >> 9; bdev = em->bdev; @@ -3130,13 +3540,6 @@ pg_offset += iosize; continue; } - /* leave this out until we have a page_mkwrite call */ - if (0 && !test_range_bit(tree, cur, cur + iosize - 1, - EXTENT_DIRTY, 0, NULL)) { - cur = cur + iosize; - pg_offset += iosize; - continue; - } if (tree->ops && tree->ops->writepage_io_hook) { ret = tree->ops->writepage_io_hook(page, cur, @@ -3147,21 +3550,20 @@ if (ret) { SetPageError(page); } else { - unsigned long max_nr = end_index + 1; + unsigned long max_nr = (i_size >> PAGE_CACHE_SHIFT) + 1; set_range_writeback(tree, cur, cur + iosize - 1); if (!PageWriteback(page)) { - printk(KERN_ERR "btrfs warning page %lu not " - "writeback, cur %llu end %llu\n", - page->index, (unsigned long long)cur, - (unsigned long long)end); + btrfs_err(BTRFS_I(inode)->root->fs_info, + "page %lu not writeback, cur %llu end %llu", + page->index, cur, end); } - ret = submit_extent_page(write_flags, tree, page, + ret = submit_extent_page(write_flags, tree, wbc, page, sector, iosize, pg_offset, bdev, &epd->bio, max_nr, end_bio_extent_writepage, - 0, 0, 0); + 0, 0, 0, false); if (ret) SetPageError(page); } @@ -3170,35 +3572,107 @@ nr++; } done: - if (nr == 0) { - /* make sure the mapping tag for page dirty gets cleared */ - set_page_writeback(page); - end_page_writeback(page); - } - unlock_page(page); + *nr_ret = nr; done_unlocked: /* drop our reference on any cached states */ free_extent_state(cached_state); - return 0; + return ret; } -static int eb_wait(void *word) +/* + * the writepage semantics are similar to regular writepage. extent + * records are inserted to lock ranges in the tree, and as dirty areas + * are found, they are marked writeback. Then the lock bits are removed + * and the end_io handler clears the writeback ranges + */ +static int __extent_writepage(struct page *page, struct writeback_control *wbc, + void *data) { - io_schedule(); + struct inode *inode = page->mapping->host; + struct extent_page_data *epd = data; + u64 start = page_offset(page); + u64 page_end = start + PAGE_CACHE_SIZE - 1; + int ret; + int nr = 0; + size_t pg_offset = 0; + loff_t i_size = i_size_read(inode); + unsigned long end_index = i_size >> PAGE_CACHE_SHIFT; + int write_flags; + unsigned long nr_written = 0; + + if (wbc->sync_mode == WB_SYNC_ALL) + write_flags = WRITE_SYNC; + else + write_flags = WRITE; + + trace___extent_writepage(page, inode, wbc); + + WARN_ON(!PageLocked(page)); + + ClearPageError(page); + + pg_offset = i_size & (PAGE_CACHE_SIZE - 1); + if (page->index > end_index || + (page->index == end_index && !pg_offset)) { + page->mapping->a_ops->invalidatepage(page, 0, PAGE_CACHE_SIZE); + unlock_page(page); + return 0; + } + + if (page->index == end_index) { + char *userpage; + + userpage = kmap_atomic(page); + memset(userpage + pg_offset, 0, + PAGE_CACHE_SIZE - pg_offset); + kunmap_atomic(userpage); + flush_dcache_page(page); + } + + pg_offset = 0; + + set_page_extent_mapped(page); + + ret = writepage_delalloc(inode, page, wbc, epd, start, &nr_written); + if (ret == 1) + goto done_unlocked; + if (ret) + goto done; + + ret = __extent_writepage_io(inode, page, wbc, epd, + i_size, nr_written, write_flags, &nr); + if (ret == 1) + goto done_unlocked; + +done: + if (nr == 0) { + /* make sure the mapping tag for page dirty gets cleared */ + set_page_writeback(page); + end_page_writeback(page); + } + if (PageError(page)) { + ret = ret < 0 ? ret : -EIO; + end_extent_writepage(page, ret, start, page_end); + } + unlock_page(page); + return ret; + +done_unlocked: return 0; } void wait_on_extent_buffer_writeback(struct extent_buffer *eb) { - wait_on_bit(&eb->bflags, EXTENT_BUFFER_WRITEBACK, eb_wait, - TASK_UNINTERRUPTIBLE); + wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_WRITEBACK, + TASK_UNINTERRUPTIBLE); } -static int lock_extent_buffer_for_io(struct extent_buffer *eb, - struct btrfs_fs_info *fs_info, - struct extent_page_data *epd) +static noinline_for_stack int +lock_extent_buffer_for_io(struct extent_buffer *eb, + struct btrfs_fs_info *fs_info, + struct extent_page_data *epd) { unsigned long i, num_pages; int flush = 0; @@ -3252,7 +3726,7 @@ num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { - struct page *p = extent_buffer_page(eb, i); + struct page *p = eb->pages[i]; if (!trylock_page(p)) { if (!flush) { @@ -3269,29 +3743,89 @@ static void end_extent_buffer_writeback(struct extent_buffer *eb) { clear_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&eb->bflags, EXTENT_BUFFER_WRITEBACK); } -static void end_bio_extent_buffer_writepage(struct bio *bio, int err) +static void set_btree_ioerr(struct page *page) { - int uptodate = err == 0; - struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; + struct extent_buffer *eb = (struct extent_buffer *)page->private; + struct btrfs_inode *btree_ino = BTRFS_I(eb->fs_info->btree_inode); + + SetPageError(page); + if (test_and_set_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags)) + return; + + /* + * If writeback for a btree extent that doesn't belong to a log tree + * failed, increment the counter transaction->eb_write_errors. + * We do this because while the transaction is running and before it's + * committing (when we call filemap_fdata[write|wait]_range against + * the btree inode), we might have + * btree_inode->i_mapping->a_ops->writepages() called by the VM - if it + * returns an error or an error happens during writeback, when we're + * committing the transaction we wouldn't know about it, since the pages + * can be no longer dirty nor marked anymore for writeback (if a + * subsequent modification to the extent buffer didn't happen before the + * transaction commit), which makes filemap_fdata[write|wait]_range not + * able to find the pages tagged with SetPageError at transaction + * commit time. So if this happens we must abort the transaction, + * otherwise we commit a super block with btree roots that point to + * btree nodes/leafs whose content on disk is invalid - either garbage + * or the content of some node/leaf from a past generation that got + * cowed or deleted and is no longer valid. + * + * Note: setting AS_EIO/AS_ENOSPC in the btree inode's i_mapping would + * not be enough - we need to distinguish between log tree extents vs + * non-log tree extents, and the next filemap_fdatawait_range() call + * will catch and clear such errors in the mapping - and that call might + * be from a log sync and not from a transaction commit. Also, checking + * for the eb flag EXTENT_BUFFER_WRITE_ERR at transaction commit time is + * not done and would not be reliable - the eb might have been released + * from memory and reading it back again means that flag would not be + * set (since it's a runtime flag, not persisted on disk). + * + * Using the flags below in the btree inode also makes us achieve the + * goal of AS_EIO/AS_ENOSPC when writepages() returns success, started + * writeback for all dirty pages and before filemap_fdatawait_range() + * is called, the writeback for all dirty pages had already finished + * with errors - because we were not using AS_EIO/AS_ENOSPC, + * filemap_fdatawait_range() would return success, as it could not know + * that writeback errors happened (the pages were no longer tagged for + * writeback). + */ + switch (eb->log_index) { + case -1: + set_bit(BTRFS_INODE_BTREE_ERR, &btree_ino->runtime_flags); + break; + case 0: + set_bit(BTRFS_INODE_BTREE_LOG1_ERR, &btree_ino->runtime_flags); + break; + case 1: + set_bit(BTRFS_INODE_BTREE_LOG2_ERR, &btree_ino->runtime_flags); + break; + default: + BUG(); /* unexpected, logic error */ + } +} + +static void end_bio_extent_buffer_writepage(struct bio *bio) +{ + struct bio_vec *bvec; struct extent_buffer *eb; - int done; + int i, done; - do { + bio_for_each_segment_all(bvec, bio, i) { struct page *page = bvec->bv_page; - bvec--; eb = (struct extent_buffer *)page->private; BUG_ON(!eb); done = atomic_dec_and_test(&eb->io_pages); - if (!uptodate || test_bit(EXTENT_BUFFER_IOERR, &eb->bflags)) { - set_bit(EXTENT_BUFFER_IOERR, &eb->bflags); + if (bio->bi_error || + test_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags)) { ClearPageUptodate(page); - SetPageError(page); + set_btree_ioerr(page); } end_page_writeback(page); @@ -3300,43 +3834,43 @@ continue; end_extent_buffer_writeback(eb); - } while (bvec >= bio->bi_io_vec); + } bio_put(bio); - } -static int write_one_eb(struct extent_buffer *eb, +static noinline_for_stack int write_one_eb(struct extent_buffer *eb, struct btrfs_fs_info *fs_info, struct writeback_control *wbc, struct extent_page_data *epd) { struct block_device *bdev = fs_info->fs_devices->latest_bdev; + struct extent_io_tree *tree = &BTRFS_I(fs_info->btree_inode)->io_tree; u64 offset = eb->start; unsigned long i, num_pages; unsigned long bio_flags = 0; int rw = (epd->sync_io ? WRITE_SYNC : WRITE) | REQ_META; int ret = 0; - clear_bit(EXTENT_BUFFER_IOERR, &eb->bflags); + clear_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags); num_pages = num_extent_pages(eb->start, eb->len); atomic_set(&eb->io_pages, num_pages); if (btrfs_header_owner(eb) == BTRFS_TREE_LOG_OBJECTID) bio_flags = EXTENT_BIO_TREE_LOG; for (i = 0; i < num_pages; i++) { - struct page *p = extent_buffer_page(eb, i); + struct page *p = eb->pages[i]; clear_page_dirty_for_io(p); set_page_writeback(p); - ret = submit_extent_page(rw, eb->tree, p, offset >> 9, + ret = submit_extent_page(rw, tree, wbc, p, offset >> 9, PAGE_CACHE_SIZE, 0, bdev, &epd->bio, -1, end_bio_extent_buffer_writepage, - 0, epd->bio_flags, bio_flags); + 0, epd->bio_flags, bio_flags, false); epd->bio_flags = bio_flags; if (ret) { - set_bit(EXTENT_BUFFER_IOERR, &eb->bflags); - SetPageError(p); + set_btree_ioerr(p); + end_page_writeback(p); if (atomic_sub_and_test(num_pages - i, &eb->io_pages)) end_extent_buffer_writeback(eb); ret = -EIO; @@ -3349,7 +3883,8 @@ if (unlikely(ret)) { for (; i < num_pages; i++) { - struct page *p = extent_buffer_page(eb, i); + struct page *p = eb->pages[i]; + clear_page_dirty_for_io(p); unlock_page(p); } } @@ -3426,9 +3961,8 @@ * but no sense in crashing the users box for something * we can survive anyway. */ - if (!eb) { + if (WARN_ON(!eb)) { spin_unlock(&mapping->private_lock); - WARN_ON(1); continue; } @@ -3504,6 +4038,7 @@ struct inode *inode = mapping->host; int ret = 0; int done = 0; + int err = 0; int nr_to_write_done = 0; struct pagevec pvec; int nr_pages; @@ -3590,8 +4125,8 @@ unlock_page(page); ret = 0; } - if (ret) - done = 1; + if (!err && ret < 0) + err = ret; /* * the filesystem may choose to bump up nr_to_write. @@ -3603,7 +4138,7 @@ pagevec_release(&pvec); cond_resched(); } - if (!scanned && !done) { + if (!scanned && !done && !err) { /* * We hit the last page and there is more work to be done: wrap * back to the start of the file @@ -3613,7 +4148,7 @@ goto retry; } btrfs_add_delayed_iput(inode); - return ret; + return err; } static void flush_epd_write_bio(struct extent_page_data *epd) @@ -3733,8 +4268,9 @@ unsigned long bio_flags = 0; struct page *pagepool[16]; struct page *page; - int i = 0; + struct extent_map *em_cached = NULL; int nr = 0; + u64 prev_em_start = (u64)-1; for (page_idx = 0; page_idx < nr_pages; page_idx++) { page = list_entry(pages->prev, struct page, lru); @@ -3750,18 +4286,16 @@ pagepool[nr++] = page; if (nr < ARRAY_SIZE(pagepool)) continue; - for (i = 0; i < nr; i++) { - __extent_read_full_page(tree, pagepool[i], get_extent, - &bio, 0, &bio_flags, READ); - page_cache_release(pagepool[i]); - } + __extent_readpages(tree, pagepool, nr, get_extent, &em_cached, + &bio, 0, &bio_flags, READ, &prev_em_start); nr = 0; } - for (i = 0; i < nr; i++) { - __extent_read_full_page(tree, pagepool[i], get_extent, - &bio, 0, &bio_flags, READ); - page_cache_release(pagepool[i]); - } + if (nr) + __extent_readpages(tree, pagepool, nr, get_extent, &em_cached, + &bio, 0, &bio_flags, READ, &prev_em_start); + + if (em_cached) + free_extent_map(em_cached); BUG_ON(!list_empty(pages)); if (bio) @@ -3846,7 +4380,7 @@ u64 start = page_offset(page); u64 end = start + PAGE_CACHE_SIZE - 1; - if ((mask & __GFP_WAIT) && + if (gfpflags_allow_blocking(mask) && page->mapping->host->i_size > 16 * 1024 * 1024) { u64 len; while (start <= end) { @@ -3897,7 +4431,7 @@ if (offset >= last) return NULL; - while(1) { + while (1) { len = last - offset; if (len == 0) break; @@ -3937,12 +4471,11 @@ struct extent_map *em = NULL; struct extent_state *cached_state = NULL; struct btrfs_path *path; - struct btrfs_file_extent_item *item; + struct btrfs_root *root = BTRFS_I(inode)->root; int end = 0; u64 em_start = 0; u64 em_len = 0; u64 em_end = 0; - unsigned long emflags; if (len == 0) return -EINVAL; @@ -3952,25 +4485,23 @@ return -ENOMEM; path->leave_spinning = 1; - start = ALIGN(start, BTRFS_I(inode)->root->sectorsize); - len = ALIGN(len, BTRFS_I(inode)->root->sectorsize); + start = round_down(start, BTRFS_I(inode)->root->sectorsize); + len = round_up(max, BTRFS_I(inode)->root->sectorsize) - start; /* * lookup the last file extent. We're not using i_size here * because there might be preallocation past i_size */ - ret = btrfs_lookup_file_extent(NULL, BTRFS_I(inode)->root, - path, btrfs_ino(inode), -1, 0); + ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), -1, + 0); if (ret < 0) { btrfs_free_path(path); return ret; } WARN_ON(!ret); path->slots[0]--; - item = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_file_extent_item); btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]); - found_type = btrfs_key_type(&found_key); + found_type = found_key.type; /* No extents, but there might be delalloc bits */ if (found_key.objectid != btrfs_ino(inode) || @@ -3987,7 +4518,7 @@ last = found_key.offset; last_for_get_extent = last + 1; } - btrfs_free_path(path); + btrfs_release_path(path); /* * we might have some extents allocated but more delalloc past those @@ -4012,7 +4543,7 @@ } while (!end) { - u64 offset_in_extent; + u64 offset_in_extent = 0; /* break if the extent we found is outside the range */ if (em->start >= max || extent_map_end(em) < off) @@ -4028,12 +4559,14 @@ /* * record the offset from the start of the extent - * for adjusting the disk offset below + * for adjusting the disk offset below. Only do this if the + * extent isn't compressed since our in ram offset may be past + * what we have actually allocated on disk. */ - offset_in_extent = em_start - em->start; + if (!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) + offset_in_extent = em_start - em->start; em_end = extent_map_end(em); em_len = em_end - em_start; - emflags = em->flags; disko = 0; flags = 0; @@ -4053,11 +4586,32 @@ } else if (em->block_start == EXTENT_MAP_DELALLOC) { flags |= (FIEMAP_EXTENT_DELALLOC | FIEMAP_EXTENT_UNKNOWN); - } else { + } else if (fieinfo->fi_extents_max) { + u64 bytenr = em->block_start - + (em->start - em->orig_start); + disko = em->block_start + offset_in_extent; + + /* + * As btrfs supports shared space, this information + * can be exported to userspace tools via + * flag FIEMAP_EXTENT_SHARED. If fi_extents_max == 0 + * then we're just getting a count and we can skip the + * lookup stuff. + */ + ret = btrfs_check_shared(NULL, root->fs_info, + root->objectid, + btrfs_ino(inode), bytenr); + if (ret < 0) + goto out_free; + if (ret) + flags |= FIEMAP_EXTENT_SHARED; + ret = 0; } if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) flags |= FIEMAP_EXTENT_ENCODED; + if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + flags |= FIEMAP_EXTENT_UNWRITTEN; free_extent_map(em); em = NULL; @@ -4089,6 +4643,7 @@ out_free: free_extent_map(em); out: + btrfs_free_path(path); unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len - 1, &cached_state, GFP_NOFS); return ret; @@ -4100,19 +4655,84 @@ kmem_cache_free(extent_buffer_cache, eb); } -static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree, - u64 start, - unsigned long len, - gfp_t mask) +int extent_buffer_under_io(struct extent_buffer *eb) +{ + return (atomic_read(&eb->io_pages) || + test_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags) || + test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); +} + +/* + * Helper for releasing extent buffer page. + */ +static void btrfs_release_extent_buffer_page(struct extent_buffer *eb) +{ + unsigned long index; + struct page *page; + int mapped = !test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags); + + BUG_ON(extent_buffer_under_io(eb)); + + index = num_extent_pages(eb->start, eb->len); + if (index == 0) + return; + + do { + index--; + page = eb->pages[index]; + if (!page) + continue; + if (mapped) + spin_lock(&page->mapping->private_lock); + /* + * We do this since we'll remove the pages after we've + * removed the eb from the radix tree, so we could race + * and have this page now attached to the new eb. So + * only clear page_private if it's still connected to + * this eb. + */ + if (PagePrivate(page) && + page->private == (unsigned long)eb) { + BUG_ON(test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); + BUG_ON(PageDirty(page)); + BUG_ON(PageWriteback(page)); + /* + * We need to make sure we haven't be attached + * to a new eb. + */ + ClearPagePrivate(page); + set_page_private(page, 0); + /* One for the page private */ + page_cache_release(page); + } + + if (mapped) + spin_unlock(&page->mapping->private_lock); + + /* One for when we alloced the page */ + page_cache_release(page); + } while (index != 0); +} + +/* + * Helper for releasing the extent buffer. + */ +static inline void btrfs_release_extent_buffer(struct extent_buffer *eb) +{ + btrfs_release_extent_buffer_page(eb); + __free_extent_buffer(eb); +} + +static struct extent_buffer * +__alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start, + unsigned long len) { struct extent_buffer *eb = NULL; - eb = kmem_cache_zalloc(extent_buffer_cache, mask); - if (eb == NULL) - return NULL; + eb = kmem_cache_zalloc(extent_buffer_cache, GFP_NOFS|__GFP_NOFAIL); eb->start = start; eb->len = len; - eb->tree = tree; + eb->fs_info = fs_info; eb->bflags = 0; rwlock_init(&eb->lock); atomic_set(&eb->write_locks, 0); @@ -4148,13 +4768,16 @@ struct extent_buffer *new; unsigned long num_pages = num_extent_pages(src->start, src->len); - new = __alloc_extent_buffer(NULL, src->start, src->len, GFP_ATOMIC); + new = __alloc_extent_buffer(src->fs_info, src->start, src->len); if (new == NULL) return NULL; for (i = 0; i < num_pages; i++) { - p = alloc_page(GFP_ATOMIC); - BUG_ON(!p); + p = alloc_page(GFP_NOFS); + if (!p) { + btrfs_release_extent_buffer(new); + return NULL; + } attach_extent_buffer_page(new, p); WARN_ON(PageDirty(p)); SetPageUptodate(p); @@ -4168,18 +4791,31 @@ return new; } -struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len) +struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info, + u64 start) { struct extent_buffer *eb; - unsigned long num_pages = num_extent_pages(0, len); + unsigned long len; + unsigned long num_pages; unsigned long i; - eb = __alloc_extent_buffer(NULL, start, len, GFP_ATOMIC); + if (!fs_info) { + /* + * Called only from tests that don't always have a fs_info + * available, but we know that nodesize is 4096 + */ + len = 4096; + } else { + len = fs_info->tree_root->nodesize; + } + num_pages = num_extent_pages(0, len); + + eb = __alloc_extent_buffer(fs_info, start, len); if (!eb) return NULL; for (i = 0; i < num_pages; i++) { - eb->pages[i] = alloc_page(GFP_ATOMIC); + eb->pages[i] = alloc_page(GFP_NOFS); if (!eb->pages[i]) goto err; } @@ -4195,76 +4831,6 @@ return NULL; } -static int extent_buffer_under_io(struct extent_buffer *eb) -{ - return (atomic_read(&eb->io_pages) || - test_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags) || - test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); -} - -/* - * Helper for releasing extent buffer page. - */ -static void btrfs_release_extent_buffer_page(struct extent_buffer *eb, - unsigned long start_idx) -{ - unsigned long index; - unsigned long num_pages; - struct page *page; - int mapped = !test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags); - - BUG_ON(extent_buffer_under_io(eb)); - - num_pages = num_extent_pages(eb->start, eb->len); - index = start_idx + num_pages; - if (start_idx >= index) - return; - - do { - index--; - page = extent_buffer_page(eb, index); - if (page && mapped) { - spin_lock(&page->mapping->private_lock); - /* - * We do this since we'll remove the pages after we've - * removed the eb from the radix tree, so we could race - * and have this page now attached to the new eb. So - * only clear page_private if it's still connected to - * this eb. - */ - if (PagePrivate(page) && - page->private == (unsigned long)eb) { - BUG_ON(test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); - BUG_ON(PageDirty(page)); - BUG_ON(PageWriteback(page)); - /* - * We need to make sure we haven't be attached - * to a new eb. - */ - ClearPagePrivate(page); - set_page_private(page, 0); - /* One for the page private */ - page_cache_release(page); - } - spin_unlock(&page->mapping->private_lock); - - } - if (page) { - /* One for when we alloced the page */ - page_cache_release(page); - } - } while (index != start_idx); -} - -/* - * Helper for releasing the extent buffer. - */ -static inline void btrfs_release_extent_buffer(struct extent_buffer *eb) -{ - btrfs_release_extent_buffer_page(eb, 0); - __free_extent_buffer(eb); -} - static void check_buffer_tree_ref(struct extent_buffer *eb) { int refs; @@ -4298,7 +4864,8 @@ spin_unlock(&eb->refs_lock); } -static void mark_extent_buffer_accessed(struct extent_buffer *eb) +static void mark_extent_buffer_accessed(struct extent_buffer *eb, + struct page *accessed) { unsigned long num_pages, i; @@ -4306,39 +4873,121 @@ num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { - struct page *p = extent_buffer_page(eb, i); - mark_page_accessed(p); + struct page *p = eb->pages[i]; + + if (p != accessed) + mark_page_accessed(p); + } +} + +struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info, + u64 start) +{ + struct extent_buffer *eb; + + rcu_read_lock(); + eb = radix_tree_lookup(&fs_info->buffer_radix, + start >> PAGE_CACHE_SHIFT); + if (eb && atomic_inc_not_zero(&eb->refs)) { + rcu_read_unlock(); + /* + * Lock our eb's refs_lock to avoid races with + * free_extent_buffer. When we get our eb it might be flagged + * with EXTENT_BUFFER_STALE and another task running + * free_extent_buffer might have seen that flag set, + * eb->refs == 2, that the buffer isn't under IO (dirty and + * writeback flags not set) and it's still in the tree (flag + * EXTENT_BUFFER_TREE_REF set), therefore being in the process + * of decrementing the extent buffer's reference count twice. + * So here we could race and increment the eb's reference count, + * clear its stale flag, mark it as dirty and drop our reference + * before the other task finishes executing free_extent_buffer, + * which would later result in an attempt to free an extent + * buffer that is dirty. + */ + if (test_bit(EXTENT_BUFFER_STALE, &eb->bflags)) { + spin_lock(&eb->refs_lock); + spin_unlock(&eb->refs_lock); + } + mark_extent_buffer_accessed(eb, NULL); + return eb; } + rcu_read_unlock(); + + return NULL; } -struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, - u64 start, unsigned long len) +#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS +struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info, + u64 start) { + struct extent_buffer *eb, *exists = NULL; + int ret; + + eb = find_extent_buffer(fs_info, start); + if (eb) + return eb; + eb = alloc_dummy_extent_buffer(fs_info, start); + if (!eb) + return NULL; + eb->fs_info = fs_info; +again: + ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM); + if (ret) + goto free_eb; + spin_lock(&fs_info->buffer_lock); + ret = radix_tree_insert(&fs_info->buffer_radix, + start >> PAGE_CACHE_SHIFT, eb); + spin_unlock(&fs_info->buffer_lock); + radix_tree_preload_end(); + if (ret == -EEXIST) { + exists = find_extent_buffer(fs_info, start); + if (exists) + goto free_eb; + else + goto again; + } + check_buffer_tree_ref(eb); + set_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags); + + /* + * We will free dummy extent buffer's if they come into + * free_extent_buffer with a ref count of 2, but if we are using this we + * want the buffers to stay in memory until we're done with them, so + * bump the ref count again. + */ + atomic_inc(&eb->refs); + return eb; +free_eb: + btrfs_release_extent_buffer(eb); + return exists; +} +#endif + +struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, + u64 start) +{ + unsigned long len = fs_info->tree_root->nodesize; unsigned long num_pages = num_extent_pages(start, len); unsigned long i; unsigned long index = start >> PAGE_CACHE_SHIFT; struct extent_buffer *eb; struct extent_buffer *exists = NULL; struct page *p; - struct address_space *mapping = tree->mapping; + struct address_space *mapping = fs_info->btree_inode->i_mapping; int uptodate = 1; int ret; - rcu_read_lock(); - eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); - if (eb && atomic_inc_not_zero(&eb->refs)) { - rcu_read_unlock(); - mark_extent_buffer_accessed(eb); + eb = find_extent_buffer(fs_info, start); + if (eb) return eb; - } - rcu_read_unlock(); - eb = __alloc_extent_buffer(tree, start, len, GFP_NOFS); + eb = __alloc_extent_buffer(fs_info, start, len); if (!eb) return NULL; for (i = 0; i < num_pages; i++, index++) { - p = find_or_create_page(mapping, index, GFP_NOFS); + p = find_or_create_page(mapping, index, GFP_NOFS|__GFP_NOFAIL); if (!p) goto free_eb; @@ -4356,9 +5005,10 @@ spin_unlock(&mapping->private_lock); unlock_page(p); page_cache_release(p); - mark_extent_buffer_accessed(exists); + mark_extent_buffer_accessed(exists, p); goto free_eb; } + exists = NULL; /* * Do this so attach doesn't complain and we need to @@ -4371,7 +5021,6 @@ attach_extent_buffer_page(eb, p); spin_unlock(&mapping->private_lock); WARN_ON(PageDirty(p)); - mark_page_accessed(p); eb->pages[i] = p; if (!PageUptodate(p)) uptodate = 0; @@ -4388,26 +5037,21 @@ if (ret) goto free_eb; - spin_lock(&tree->buffer_lock); - ret = radix_tree_insert(&tree->buffer, start >> PAGE_CACHE_SHIFT, eb); + spin_lock(&fs_info->buffer_lock); + ret = radix_tree_insert(&fs_info->buffer_radix, + start >> PAGE_CACHE_SHIFT, eb); + spin_unlock(&fs_info->buffer_lock); + radix_tree_preload_end(); if (ret == -EEXIST) { - exists = radix_tree_lookup(&tree->buffer, - start >> PAGE_CACHE_SHIFT); - if (!atomic_inc_not_zero(&exists->refs)) { - spin_unlock(&tree->buffer_lock); - radix_tree_preload_end(); - exists = NULL; + exists = find_extent_buffer(fs_info, start); + if (exists) + goto free_eb; + else goto again; - } - spin_unlock(&tree->buffer_lock); - radix_tree_preload_end(); - mark_extent_buffer_accessed(exists); - goto free_eb; } /* add one reference for the tree */ check_buffer_tree_ref(eb); - spin_unlock(&tree->buffer_lock); - radix_tree_preload_end(); + set_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags); /* * there is a race where release page may have @@ -4420,7 +5064,7 @@ */ SetPageChecked(eb->pages[0]); for (i = 1; i < num_pages; i++) { - p = extent_buffer_page(eb, i); + p = eb->pages[i]; ClearPageChecked(p); unlock_page(p); } @@ -4428,33 +5072,16 @@ return eb; free_eb: + WARN_ON(!atomic_dec_and_test(&eb->refs)); for (i = 0; i < num_pages; i++) { if (eb->pages[i]) unlock_page(eb->pages[i]); } - WARN_ON(!atomic_dec_and_test(&eb->refs)); btrfs_release_extent_buffer(eb); return exists; } -struct extent_buffer *find_extent_buffer(struct extent_io_tree *tree, - u64 start, unsigned long len) -{ - struct extent_buffer *eb; - - rcu_read_lock(); - eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); - if (eb && atomic_inc_not_zero(&eb->refs)) { - rcu_read_unlock(); - mark_extent_buffer_accessed(eb); - return eb; - } - rcu_read_unlock(); - - return NULL; -} - static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head) { struct extent_buffer *eb = @@ -4468,21 +5095,27 @@ { WARN_ON(atomic_read(&eb->refs) == 0); if (atomic_dec_and_test(&eb->refs)) { - if (test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags)) { - spin_unlock(&eb->refs_lock); - } else { - struct extent_io_tree *tree = eb->tree; + if (test_and_clear_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags)) { + struct btrfs_fs_info *fs_info = eb->fs_info; spin_unlock(&eb->refs_lock); - spin_lock(&tree->buffer_lock); - radix_tree_delete(&tree->buffer, + spin_lock(&fs_info->buffer_lock); + radix_tree_delete(&fs_info->buffer_radix, eb->start >> PAGE_CACHE_SHIFT); - spin_unlock(&tree->buffer_lock); + spin_unlock(&fs_info->buffer_lock); + } else { + spin_unlock(&eb->refs_lock); } /* Should be safe to release our pages at this point */ - btrfs_release_extent_buffer_page(eb, 0); + btrfs_release_extent_buffer_page(eb); +#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS + if (unlikely(test_bit(EXTENT_BUFFER_DUMMY, &eb->bflags))) { + __free_extent_buffer(eb); + return 1; + } +#endif call_rcu(&eb->rcu_head, btrfs_release_extent_buffer_rcu); return 1; } @@ -4548,7 +5181,7 @@ num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; if (!PageDirty(page)) continue; @@ -4584,7 +5217,7 @@ WARN_ON(!test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags)); for (i = 0; i < num_pages; i++) - set_page_dirty(extent_buffer_page(eb, i)); + set_page_dirty(eb->pages[i]); return was_dirty; } @@ -4597,7 +5230,7 @@ clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; if (page) ClearPageUptodate(page); } @@ -4613,7 +5246,7 @@ set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; SetPageUptodate(page); } return 0; @@ -4653,7 +5286,7 @@ num_pages = num_extent_pages(eb->start, eb->len); for (i = start_i; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; if (wait == WAIT_NONE) { if (!trylock_page(page)) goto unlock_exit; @@ -4681,11 +5314,11 @@ goto unlock_exit; } - clear_bit(EXTENT_BUFFER_IOERR, &eb->bflags); + clear_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags); eb->read_mirror = 0; atomic_set(&eb->io_pages, num_reads); for (i = start_i; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; if (!PageUptodate(page)) { ClearPageError(page); err = __extent_read_full_page(tree, page, @@ -4710,7 +5343,7 @@ return ret; for (i = start_i; i < num_pages; i++) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; wait_on_page_locked(page); if (!PageUptodate(page)) ret = -EIO; @@ -4721,7 +5354,7 @@ unlock_exit: i = start_i; while (locked_pages > 0) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; i++; unlock_page(page); locked_pages--; @@ -4744,10 +5377,10 @@ WARN_ON(start > eb->len); WARN_ON(start + len > eb->start + eb->len); - offset = (start_offset + start) & ((unsigned long)PAGE_CACHE_SIZE - 1); + offset = (start_offset + start) & (PAGE_CACHE_SIZE - 1); while (len > 0) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; cur = min(len, (PAGE_CACHE_SIZE - offset)); kaddr = page_address(page); @@ -4760,6 +5393,43 @@ } } +int read_extent_buffer_to_user(struct extent_buffer *eb, void __user *dstv, + unsigned long start, + unsigned long len) +{ + size_t cur; + size_t offset; + struct page *page; + char *kaddr; + char __user *dst = (char __user *)dstv; + size_t start_offset = eb->start & ((u64)PAGE_CACHE_SIZE - 1); + unsigned long i = (start_offset + start) >> PAGE_CACHE_SHIFT; + int ret = 0; + + WARN_ON(start > eb->len); + WARN_ON(start + len > eb->start + eb->len); + + offset = (start_offset + start) & (PAGE_CACHE_SIZE - 1); + + while (len > 0) { + page = eb->pages[i]; + + cur = min(len, (PAGE_CACHE_SIZE - offset)); + kaddr = page_address(page); + if (copy_to_user(dst, kaddr + offset, cur)) { + ret = -EFAULT; + break; + } + + dst += cur; + len -= cur; + offset = 0; + i++; + } + + return ret; +} + int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start, unsigned long min_len, char **map, unsigned long *map_start, @@ -4786,12 +5456,12 @@ if (start + min_len > eb->len) { WARN(1, KERN_ERR "btrfs bad mapping eb start %llu len %lu, " - "wanted %lu %lu\n", (unsigned long long)eb->start, - eb->len, start, min_len); + "wanted %lu %lu\n", + eb->start, eb->len, start, min_len); return -EINVAL; } - p = extent_buffer_page(eb, i); + p = eb->pages[i]; kaddr = page_address(p); *map = kaddr + offset; *map_len = PAGE_CACHE_SIZE - offset; @@ -4814,10 +5484,10 @@ WARN_ON(start > eb->len); WARN_ON(start + len > eb->start + eb->len); - offset = (start_offset + start) & ((unsigned long)PAGE_CACHE_SIZE - 1); + offset = (start_offset + start) & (PAGE_CACHE_SIZE - 1); while (len > 0) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; cur = min(len, (PAGE_CACHE_SIZE - offset)); @@ -4848,10 +5518,10 @@ WARN_ON(start > eb->len); WARN_ON(start + len > eb->start + eb->len); - offset = (start_offset + start) & ((unsigned long)PAGE_CACHE_SIZE - 1); + offset = (start_offset + start) & (PAGE_CACHE_SIZE - 1); while (len > 0) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; WARN_ON(!PageUptodate(page)); cur = min(len, PAGE_CACHE_SIZE - offset); @@ -4878,10 +5548,10 @@ WARN_ON(start > eb->len); WARN_ON(start + len > eb->start + eb->len); - offset = (start_offset + start) & ((unsigned long)PAGE_CACHE_SIZE - 1); + offset = (start_offset + start) & (PAGE_CACHE_SIZE - 1); while (len > 0) { - page = extent_buffer_page(eb, i); + page = eb->pages[i]; WARN_ON(!PageUptodate(page)); cur = min(len, PAGE_CACHE_SIZE - offset); @@ -4909,10 +5579,10 @@ WARN_ON(src->len != dst_len); offset = (start_offset + dst_offset) & - ((unsigned long)PAGE_CACHE_SIZE - 1); + (PAGE_CACHE_SIZE - 1); while (len > 0) { - page = extent_buffer_page(dst, i); + page = dst->pages[i]; WARN_ON(!PageUptodate(page)); cur = min(len, (unsigned long)(PAGE_CACHE_SIZE - offset)); @@ -4927,23 +5597,6 @@ } } -static void move_pages(struct page *dst_page, struct page *src_page, - unsigned long dst_off, unsigned long src_off, - unsigned long len) -{ - char *dst_kaddr = page_address(dst_page); - if (dst_page == src_page) { - memmove(dst_kaddr + dst_off, dst_kaddr + src_off, len); - } else { - char *src_kaddr = page_address(src_page); - char *p = dst_kaddr + dst_off + len; - char *s = src_kaddr + src_off + len; - - while (len--) - *--p = *--s; - } -} - static inline bool areas_overlap(unsigned long src, unsigned long dst, unsigned long len) { unsigned long distance = (src > dst) ? src - dst : dst - src; @@ -4983,21 +5636,23 @@ unsigned long src_i; if (src_offset + len > dst->len) { - printk(KERN_ERR "btrfs memmove bogus src_offset %lu move " - "len %lu dst len %lu\n", src_offset, len, dst->len); + btrfs_err(dst->fs_info, + "memmove bogus src_offset %lu move " + "len %lu dst len %lu", src_offset, len, dst->len); BUG_ON(1); } if (dst_offset + len > dst->len) { - printk(KERN_ERR "btrfs memmove bogus dst_offset %lu move " - "len %lu dst len %lu\n", dst_offset, len, dst->len); + btrfs_err(dst->fs_info, + "memmove bogus dst_offset %lu move " + "len %lu dst len %lu", dst_offset, len, dst->len); BUG_ON(1); } while (len > 0) { dst_off_in_page = (start_offset + dst_offset) & - ((unsigned long)PAGE_CACHE_SIZE - 1); + (PAGE_CACHE_SIZE - 1); src_off_in_page = (start_offset + src_offset) & - ((unsigned long)PAGE_CACHE_SIZE - 1); + (PAGE_CACHE_SIZE - 1); dst_i = (start_offset + dst_offset) >> PAGE_CACHE_SHIFT; src_i = (start_offset + src_offset) >> PAGE_CACHE_SHIFT; @@ -5007,8 +5662,7 @@ cur = min_t(unsigned long, cur, (unsigned long)(PAGE_CACHE_SIZE - dst_off_in_page)); - copy_pages(extent_buffer_page(dst, dst_i), - extent_buffer_page(dst, src_i), + copy_pages(dst->pages[dst_i], dst->pages[src_i], dst_off_in_page, src_off_in_page, cur); src_offset += cur; @@ -5030,13 +5684,13 @@ unsigned long src_i; if (src_offset + len > dst->len) { - printk(KERN_ERR "btrfs memmove bogus src_offset %lu move " - "len %lu len %lu\n", src_offset, len, dst->len); + btrfs_err(dst->fs_info, "memmove bogus src_offset %lu move " + "len %lu len %lu", src_offset, len, dst->len); BUG_ON(1); } if (dst_offset + len > dst->len) { - printk(KERN_ERR "btrfs memmove bogus dst_offset %lu move " - "len %lu len %lu\n", dst_offset, len, dst->len); + btrfs_err(dst->fs_info, "memmove bogus dst_offset %lu move " + "len %lu len %lu", dst_offset, len, dst->len); BUG_ON(1); } if (dst_offset < src_offset) { @@ -5048,14 +5702,13 @@ src_i = (start_offset + src_end) >> PAGE_CACHE_SHIFT; dst_off_in_page = (start_offset + dst_end) & - ((unsigned long)PAGE_CACHE_SIZE - 1); + (PAGE_CACHE_SIZE - 1); src_off_in_page = (start_offset + src_end) & - ((unsigned long)PAGE_CACHE_SIZE - 1); + (PAGE_CACHE_SIZE - 1); cur = min_t(unsigned long, len, src_off_in_page + 1); cur = min(cur, dst_off_in_page + 1); - move_pages(extent_buffer_page(dst, dst_i), - extent_buffer_page(dst, src_i), + copy_pages(dst->pages[dst_i], dst->pages[src_i], dst_off_in_page - cur + 1, src_off_in_page - cur + 1, cur);