summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJosef Bacik <jbacik@fusionio.com>2013-08-29 13:57:21 -0400
committerChris Mason <chris.mason@fusionio.com>2013-09-01 08:16:34 -0400
commit77cef2ec5484564eca6bd12a2b4a1e88fd766fbc (patch)
treeedc2b57493ccebe4dc4ad48daf03a2ac25f24a10
parentb12d6869f67a95692017d26313ea5736d4043d0f (diff)
Btrfs: allow partial ordered extent completion
We currently have this problem where you can truncate pages that have not yet been written for an ordered extent. We do this because the truncate will be coming behind to clean us up anyway so what's the harm right? Well if truncate fails for whatever reason we leave an orphan item around for the file to be cleaned up later. But if the user goes and truncates up the file and tries to read from the area that had been discarded previously they will get a csum error because we never actually wrote that data out. This patch fixes this by allowing us to either discard the ordered extent completely, by which I mean we just free up the space we had allocated and not add the file extent, or adjust the length of the file extent we write. We do this by setting the length we truncated down to in the ordered extent, and then we set the file extent length and ram bytes to this length. The total disk space stays unchanged since we may be compressed and we can't just chop off the disk space, but at least this way the file extent only points to the valid data. Then when the file extent is free'd the extent and csums will be freed normally. This patch is needed for the next series which will give us more graceful recovery of failed truncates. Thanks, Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
-rw-r--r--fs/btrfs/inode.c60
-rw-r--r--fs/btrfs/ordered-data.c13
-rw-r--r--fs/btrfs/ordered-data.h7
3 files changed, 62 insertions, 18 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 4d4e2de3e879..f0e41b840739 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2562,8 +2562,10 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
struct extent_state *cached_state = NULL;
struct new_sa_defrag_extent *new = NULL;
int compress_type = 0;
- int ret;
+ int ret = 0;
+ u64 logical_len = ordered_extent->len;
bool nolock;
+ bool truncated = false;
nolock = btrfs_is_free_space_inode(inode);
@@ -2572,6 +2574,14 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
goto out;
}
+ if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) {
+ truncated = true;
+ logical_len = ordered_extent->truncated_len;
+ /* Truncated the entire extent, don't bother adding */
+ if (!logical_len)
+ goto out;
+ }
+
if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
BUG_ON(!list_empty(&ordered_extent->list)); /* Logic error */
btrfs_ordered_update_i_size(inode, 0, ordered_extent);
@@ -2627,15 +2637,14 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
ret = btrfs_mark_extent_written(trans, inode,
ordered_extent->file_offset,
ordered_extent->file_offset +
- ordered_extent->len);
+ logical_len);
} else {
BUG_ON(root == root->fs_info->tree_root);
ret = insert_reserved_file_extent(trans, inode,
ordered_extent->file_offset,
ordered_extent->start,
ordered_extent->disk_len,
- ordered_extent->len,
- ordered_extent->len,
+ logical_len, logical_len,
compress_type, 0, 0,
BTRFS_FILE_EXTENT_REG);
}
@@ -2667,17 +2676,27 @@ out:
if (trans)
btrfs_end_transaction(trans, root);
- if (ret) {
- clear_extent_uptodate(io_tree, ordered_extent->file_offset,
- ordered_extent->file_offset +
- ordered_extent->len - 1, NULL, GFP_NOFS);
+ if (ret || truncated) {
+ u64 start, end;
+
+ if (truncated)
+ start = ordered_extent->file_offset + logical_len;
+ else
+ start = ordered_extent->file_offset;
+ end = ordered_extent->file_offset + ordered_extent->len - 1;
+ clear_extent_uptodate(io_tree, start, end, NULL, GFP_NOFS);
+
+ /* Drop the cache for the part of the extent we didn't write. */
+ btrfs_drop_extent_cache(inode, start, end, 0);
/*
* If the ordered extent had an IOERR or something else went
* wrong we need to return the space for this ordered extent
- * back to the allocator.
+ * back to the allocator. We only free the extent in the
+ * truncated case if we didn't write out the extent at all.
*/
- if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
+ if ((ret || !logical_len) &&
+ !test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
!test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags))
btrfs_free_reserved_extent(root, ordered_extent->start,
ordered_extent->disk_len);
@@ -7336,10 +7355,23 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset,
* whoever cleared the private bit is responsible
* for the finish_ordered_io
*/
- if (TestClearPagePrivate2(page) &&
- btrfs_dec_test_ordered_pending(inode, &ordered, page_start,
- PAGE_CACHE_SIZE, 1)) {
- btrfs_finish_ordered_io(ordered);
+ if (TestClearPagePrivate2(page)) {
+ struct btrfs_ordered_inode_tree *tree;
+ u64 new_len;
+
+ tree = &BTRFS_I(inode)->ordered_tree;
+
+ spin_lock_irq(&tree->lock);
+ set_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags);
+ new_len = page_start - ordered->file_offset;
+ if (new_len < ordered->truncated_len)
+ ordered->truncated_len = new_len;
+ spin_unlock_irq(&tree->lock);
+
+ if (btrfs_dec_test_ordered_pending(inode, &ordered,
+ page_start,
+ PAGE_CACHE_SIZE, 1))
+ btrfs_finish_ordered_io(ordered);
}
btrfs_put_ordered_extent(ordered);
cached_state = NULL;
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index f2023ccb7cf6..966b413a33b8 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -205,6 +205,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
entry->bytes_left = len;
entry->inode = igrab(inode);
entry->compress_type = compress_type;
+ entry->truncated_len = (u64)-1;
if (type != BTRFS_ORDERED_IO_DONE && type != BTRFS_ORDERED_COMPLETE)
set_bit(type, &entry->flags);
@@ -920,12 +921,16 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
struct btrfs_ordered_extent *test;
int ret = 1;
- if (ordered)
+ spin_lock_irq(&tree->lock);
+ if (ordered) {
offset = entry_end(ordered);
- else
+ if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags))
+ offset = min(offset,
+ ordered->file_offset +
+ ordered->truncated_len);
+ } else {
offset = ALIGN(offset, BTRFS_I(inode)->root->sectorsize);
-
- spin_lock_irq(&tree->lock);
+ }
disk_i_size = BTRFS_I(inode)->disk_i_size;
/* truncate file */
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 68844d59ee6f..d9a5aa097b4f 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -69,6 +69,7 @@ struct btrfs_ordered_sum {
* the isize. */
#define BTRFS_ORDERED_LOGGED_CSUM 8 /* We've logged the csums on this ordered
ordered extent */
+#define BTRFS_ORDERED_TRUNCATED 9 /* Set when we have to truncate an extent */
struct btrfs_ordered_extent {
/* logical offset in the file */
@@ -96,6 +97,12 @@ struct btrfs_ordered_extent {
*/
u64 outstanding_isize;
+ /*
+ * If we get truncated we need to adjust the file extent we enter for
+ * this ordered extent so that we do not expose stale data.
+ */
+ u64 truncated_len;
+
/* flags (described above) */
unsigned long flags;