int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode,
struct btrfs_path *path, u64 start, u64 end,
- u64 *drop_end, int drop_cache)
+ u64 *drop_end, int drop_cache,
+ int replace_extent,
+ u32 extent_item_size,
+ int *key_inserted)
{
struct extent_buffer *leaf;
struct btrfs_file_extent_item *fi;
int modify_tree = -1;
int update_refs = (root->ref_cows || root == root->fs_info->tree_root);
int found = 0;
+ int leafs_visited = 0;
if (drop_cache)
btrfs_drop_extent_cache(inode, start, end - 1, 0);
path->slots[0]--;
}
ret = 0;
+ leafs_visited++;
next_slot:
leaf = path->nodes[0];
if (path->slots[0] >= btrfs_header_nritems(leaf)) {
ret = 0;
break;
}
+ leafs_visited++;
leaf = path->nodes[0];
recow = 1;
}
btrfs_file_extent_num_bytes(leaf, fi);
} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
extent_end = key.offset +
- btrfs_file_extent_inline_len(leaf, fi);
+ btrfs_file_extent_inline_len(leaf,
+ path->slots[0], fi);
} else {
WARN_ON(1);
extent_end = search_start;
}
if (!ret && del_nr > 0) {
+ /*
+ * Set path->slots[0] to first slot, so that after the delete
+ * if items are move off from our leaf to its immediate left or
+ * right neighbor leafs, we end up with a correct and adjusted
+ * path->slots[0] for our insertion.
+ */
+ path->slots[0] = del_slot;
ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
if (ret)
btrfs_abort_transaction(trans, root, ret);
+
+ leaf = path->nodes[0];
+ /*
+ * leaf eb has flag EXTENT_BUFFER_STALE if it was deleted (that
+ * is, its contents got pushed to its neighbors), in which case
+ * it means path->locks[0] == 0
+ */
+ if (!ret && replace_extent && leafs_visited == 1 &&
+ path->locks[0] &&
+ btrfs_leaf_free_space(root, leaf) >=
+ sizeof(struct btrfs_item) + extent_item_size) {
+
+ key.objectid = ino;
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = start;
+ setup_items_for_insert(root, path, &key,
+ &extent_item_size,
+ extent_item_size,
+ sizeof(struct btrfs_item) +
+ extent_item_size, 1);
+ *key_inserted = 1;
+ }
}
+ if (!replace_extent || !(*key_inserted))
+ btrfs_release_path(path);
if (drop_end)
*drop_end = found ? min(end, extent_end) : end;
- btrfs_release_path(path);
return ret;
}
if (!path)
return -ENOMEM;
ret = __btrfs_drop_extents(trans, root, inode, path, start, end, NULL,
- drop_cache);
+ drop_cache, 0, 0, NULL);
btrfs_free_path(path);
return ret;
}
}
/*
- * this gets pages into the page cache and locks them down, it also properly
- * waits for data=ordered extents to finish before allowing the pages to be
- * modified.
+ * this just gets pages into the page cache and locks them down.
*/
-static noinline int prepare_pages(struct btrfs_root *root, struct file *file,
- struct page **pages, size_t num_pages,
- loff_t pos, unsigned long first_index,
- size_t write_bytes, bool force_uptodate)
+static noinline int prepare_pages(struct inode *inode, struct page **pages,
+ size_t num_pages, loff_t pos,
+ size_t write_bytes, bool force_uptodate)
{
- struct extent_state *cached_state = NULL;
int i;
unsigned long index = pos >> PAGE_CACHE_SHIFT;
- struct inode *inode = file_inode(file);
gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
int err = 0;
- int faili = 0;
- u64 start_pos;
- u64 last_pos;
+ int faili;
- start_pos = pos & ~((u64)root->sectorsize - 1);
- last_pos = ((u64)index + num_pages) << PAGE_CACHE_SHIFT;
-
-again:
for (i = 0; i < num_pages; i++) {
pages[i] = find_or_create_page(inode->i_mapping, index + i,
mask | __GFP_WRITE);
}
wait_on_page_writeback(pages[i]);
}
- faili = num_pages - 1;
- err = 0;
+
+ return 0;
+fail:
+ while (faili >= 0) {
+ unlock_page(pages[faili]);
+ page_cache_release(pages[faili]);
+ faili--;
+ }
+ return err;
+
+}
+
+/*
+ * This function locks the extent and properly waits for data=ordered extents
+ * to finish before allowing the pages to be modified if need.
+ *
+ * The return value:
+ * 1 - the extent is locked
+ * 0 - the extent is not locked, and everything is OK
+ * -EAGAIN - need re-prepare the pages
+ * the other < 0 number - Something wrong happens
+ */
+static noinline int
+lock_and_cleanup_extent_if_need(struct inode *inode, struct page **pages,
+ size_t num_pages, loff_t pos,
+ u64 *lockstart, u64 *lockend,
+ struct extent_state **cached_state)
+{
+ u64 start_pos;
+ u64 last_pos;
+ int i;
+ int ret = 0;
+
+ start_pos = pos & ~((u64)PAGE_CACHE_SIZE - 1);
+ last_pos = start_pos + ((u64)num_pages << PAGE_CACHE_SHIFT) - 1;
+
if (start_pos < inode->i_size) {
struct btrfs_ordered_extent *ordered;
lock_extent_bits(&BTRFS_I(inode)->io_tree,
- start_pos, last_pos - 1, 0, &cached_state);
- ordered = btrfs_lookup_first_ordered_extent(inode,
- last_pos - 1);
+ start_pos, last_pos, 0, cached_state);
+ ordered = btrfs_lookup_first_ordered_extent(inode, last_pos);
if (ordered &&
ordered->file_offset + ordered->len > start_pos &&
- ordered->file_offset < last_pos) {
+ ordered->file_offset <= last_pos) {
btrfs_put_ordered_extent(ordered);
unlock_extent_cached(&BTRFS_I(inode)->io_tree,
- start_pos, last_pos - 1,
- &cached_state, GFP_NOFS);
+ start_pos, last_pos,
+ cached_state, GFP_NOFS);
for (i = 0; i < num_pages; i++) {
unlock_page(pages[i]);
page_cache_release(pages[i]);
}
- err = btrfs_wait_ordered_range(inode, start_pos,
- last_pos - start_pos);
- if (err)
- goto fail;
- goto again;
+ ret = btrfs_wait_ordered_range(inode, start_pos,
+ last_pos - start_pos + 1);
+ if (ret)
+ return ret;
+ else
+ return -EAGAIN;
}
if (ordered)
btrfs_put_ordered_extent(ordered);
clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos,
- last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
+ last_pos, EXTENT_DIRTY | EXTENT_DELALLOC |
EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
- 0, 0, &cached_state, GFP_NOFS);
- unlock_extent_cached(&BTRFS_I(inode)->io_tree,
- start_pos, last_pos - 1, &cached_state,
- GFP_NOFS);
+ 0, 0, cached_state, GFP_NOFS);
+ *lockstart = start_pos;
+ *lockend = last_pos;
+ ret = 1;
}
+
for (i = 0; i < num_pages; i++) {
if (clear_page_dirty_for_io(pages[i]))
account_page_redirty(pages[i]);
set_page_extent_mapped(pages[i]);
WARN_ON(!PageLocked(pages[i]));
}
- return 0;
-fail:
- while (faili >= 0) {
- unlock_page(pages[faili]);
- page_cache_release(pages[faili]);
- faili--;
- }
- return err;
+ return ret;
}
static noinline int check_can_nocow(struct inode *inode, loff_t pos,
struct inode *inode = file_inode(file);
struct btrfs_root *root = BTRFS_I(inode)->root;
struct page **pages = NULL;
+ struct extent_state *cached_state = NULL;
u64 release_bytes = 0;
+ u64 lockstart;
+ u64 lockend;
unsigned long first_index;
size_t num_written = 0;
int nrptrs;
int ret = 0;
bool only_release_metadata = false;
bool force_page_uptodate = false;
+ bool need_unlock;
nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) /
PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
}
release_bytes = reserve_bytes;
-
+ need_unlock = false;
+again:
/*
* This is going to setup the pages array with the number of
* pages we want, so we don't really need to worry about the
* contents of pages from loop to loop
*/
- ret = prepare_pages(root, file, pages, num_pages,
- pos, first_index, write_bytes,
+ ret = prepare_pages(inode, pages, num_pages,
+ pos, write_bytes,
force_page_uptodate);
if (ret)
break;
+ ret = lock_and_cleanup_extent_if_need(inode, pages, num_pages,
+ pos, &lockstart, &lockend,
+ &cached_state);
+ if (ret < 0) {
+ if (ret == -EAGAIN)
+ goto again;
+ break;
+ } else if (ret > 0) {
+ need_unlock = true;
+ ret = 0;
+ }
+
copied = btrfs_copy_from_user(pos, num_pages,
write_bytes, pages, i);
}
release_bytes = dirty_pages << PAGE_CACHE_SHIFT;
- if (copied > 0) {
+
+ if (copied > 0)
ret = btrfs_dirty_pages(root, inode, pages,
dirty_pages, pos, copied,
NULL);
- if (ret) {
- btrfs_drop_pages(pages, num_pages);
- break;
- }
+ if (need_unlock)
+ unlock_extent_cached(&BTRFS_I(inode)->io_tree,
+ lockstart, lockend, &cached_state,
+ GFP_NOFS);
+ if (ret) {
+ btrfs_drop_pages(pages, num_pages);
+ break;
}
release_bytes = 0;
- btrfs_drop_pages(pages, num_pages);
-
if (only_release_metadata && copied > 0) {
u64 lockstart = round_down(pos, root->sectorsize);
u64 lockend = lockstart +
only_release_metadata = false;
}
+ btrfs_drop_pages(pages, num_pages);
+
cond_resched();
balance_dirty_pages_ratelimited(inode->i_mapping);
if (file->private_data)
btrfs_ioctl_trans_end(file);
+ /*
+ * We use start here because we will need to wait on the IO to complete
+ * in btrfs_sync_log, which could require joining a transaction (for
+ * example checking cross references in the nocow path). If we use join
+ * here we could get into a situation where we're waiting on IO to
+ * happen that is blocked on a transaction trying to commit. With start
+ * we inc the extwriter counter, so we wait for all extwriters to exit
+ * before we start blocking join'ers. This comment is to keep somebody
+ * from thinking they are super smart and changing this to
+ * btrfs_join_transaction *cough*Josef*cough*.
+ */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
mutex_unlock(&inode->i_mutex);
goto out;
}
+ trans->sync = true;
ret = btrfs_log_dentry_safe(trans, root, dentry);
if (ret < 0) {
* we need to try again.
*/
if ((!ordered ||
- (ordered->file_offset + ordered->len < lockstart ||
+ (ordered->file_offset + ordered->len <= lockstart ||
ordered->file_offset > lockend)) &&
!test_range_bit(&BTRFS_I(inode)->io_tree, lockstart,
lockend, EXTENT_UPTODATE, 0,
while (cur_offset < lockend) {
ret = __btrfs_drop_extents(trans, root, inode, path,
cur_offset, lockend + 1,
- &drop_end, 1);
+ &drop_end, 1, 0, 0, NULL);
if (ret != -ENOSPC)
break;
projects / ~andy / linux / blobdiff