diff mbox series

[06/16] btrfs: Lock extents before pages in writepages

Message ID 28cb7dbc0216d2a5f55efd296113f9f9576dda41.1668530684.git.rgoldwyn@suse.com (mailing list archive)
State New, archived
Headers show
Series Lock extents before pages | expand

Commit Message

Goldwyn Rodrigues Nov. 15, 2022, 6 p.m. UTC
writepages() locks the extents in find_lock_delalloc_range() and unlocks
using clear_bit EXTENT_LOCKED operations is cow/delalloc operations.

Call extent locking/unlocking around writepages() sequence as opposed to
while performing delayed allocation.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/extent_io.c |  5 -----
 fs/btrfs/inode.c     | 43 +++++++++++++++++++++++++++++++------------
 2 files changed, 31 insertions(+), 17 deletions(-)

Comments

Josef Bacik Dec. 13, 2022, 6:39 p.m. UTC | #1
On Tue, Nov 15, 2022 at 12:00:24PM -0600, Goldwyn Rodrigues wrote:
> writepages() locks the extents in find_lock_delalloc_range() and unlocks
> using clear_bit EXTENT_LOCKED operations is cow/delalloc operations.
> 
> Call extent locking/unlocking around writepages() sequence as opposed to
> while performing delayed allocation.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> ---
>  fs/btrfs/extent_io.c |  5 -----
>  fs/btrfs/inode.c     | 43 +++++++++++++++++++++++++++++++------------
>  2 files changed, 31 insertions(+), 17 deletions(-)
> 
> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
> index 92068e4ff9c3..42bae149f923 100644
> --- a/fs/btrfs/extent_io.c
> +++ b/fs/btrfs/extent_io.c
> @@ -464,15 +464,10 @@ noinline_for_stack bool find_lock_delalloc_range(struct inode *inode,
>  		}
>  	}
>  
> -	/* step three, lock the state bits for the whole range */
> -	lock_extent(tree, delalloc_start, delalloc_end, &cached_state);
> -
>  	/* then test to make sure it is all still delalloc */
>  	ret = test_range_bit(tree, delalloc_start, delalloc_end,
>  			     EXTENT_DELALLOC, 1, cached_state);
>  	if (!ret) {
> -		unlock_extent(tree, delalloc_start, delalloc_end,
> -			      &cached_state);
>  		__unlock_for_delalloc(inode, locked_page,
>  			      delalloc_start, delalloc_end);
>  		cond_resched();
> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> index 4bfa51871ddc..92726831dd5d 100644
> --- a/fs/btrfs/inode.c
> +++ b/fs/btrfs/inode.c
> @@ -992,7 +992,6 @@ static int submit_one_async_extent(struct btrfs_inode *inode,
>  				   struct async_extent *async_extent,
>  				   u64 *alloc_hint)
>  {
> -	struct extent_io_tree *io_tree = &inode->io_tree;
>  	struct btrfs_root *root = inode->root;
>  	struct btrfs_fs_info *fs_info = root->fs_info;
>  	struct btrfs_key ins;
> @@ -1013,7 +1012,6 @@ static int submit_one_async_extent(struct btrfs_inode *inode,
>  		if (!(start >= locked_page_end || end <= locked_page_start))
>  			locked_page = async_chunk->locked_page;
>  	}
> -	lock_extent(io_tree, start, end, NULL);
>  
>  	/* We have fall back to uncompressed write */
>  	if (!async_extent->pages)
> @@ -1067,7 +1065,7 @@ static int submit_one_async_extent(struct btrfs_inode *inode,
>  
>  	/* Clear dirty, set writeback and unlock the pages. */
>  	extent_clear_unlock_delalloc(inode, start, end,
> -			NULL, EXTENT_LOCKED | EXTENT_DELALLOC,
> +			NULL, EXTENT_DELALLOC,
>  			PAGE_UNLOCK | PAGE_START_WRITEBACK);
>  	if (btrfs_submit_compressed_write(inode, start,	/* file_offset */
>  			    async_extent->ram_size,	/* num_bytes */
> @@ -1095,7 +1093,7 @@ static int submit_one_async_extent(struct btrfs_inode *inode,
>  	btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1);
>  out_free:
>  	extent_clear_unlock_delalloc(inode, start, end,
> -				     NULL, EXTENT_LOCKED | EXTENT_DELALLOC |
> +				     NULL, EXTENT_DELALLOC |
>  				     EXTENT_DELALLOC_NEW |
>  				     EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING,
>  				     PAGE_UNLOCK | PAGE_START_WRITEBACK |
> @@ -1263,7 +1261,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode,
>  			 */
>  			extent_clear_unlock_delalloc(inode, start, end,
>  				     locked_page,
> -				     EXTENT_LOCKED | EXTENT_DELALLOC |
> +				     EXTENT_DELALLOC |
>  				     EXTENT_DELALLOC_NEW | EXTENT_DEFRAG |
>  				     EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
>  				     PAGE_START_WRITEBACK | PAGE_END_WRITEBACK);
> @@ -1374,7 +1372,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode,
>  
>  		extent_clear_unlock_delalloc(inode, start, start + ram_size - 1,
>  					     locked_page,
> -					     EXTENT_LOCKED | EXTENT_DELALLOC,
> +					     EXTENT_DELALLOC,
>  					     page_ops);
>  		if (num_bytes < cur_alloc_size)
>  			num_bytes = 0;
> @@ -1425,7 +1423,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode,
>  	 * We process each region below.
>  	 */
>  
> -	clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
> +	clear_bits = EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
>  		EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV;
>  	page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK;
>  
> @@ -1575,7 +1573,7 @@ static int cow_file_range_async(struct btrfs_inode *inode,
>  	memalloc_nofs_restore(nofs_flag);
>  
>  	if (!ctx) {
> -		unsigned clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC |
> +		unsigned clear_bits = EXTENT_DELALLOC |
>  			EXTENT_DELALLOC_NEW | EXTENT_DEFRAG |
>  			EXTENT_DO_ACCOUNTING;
>  		unsigned long page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK |
> @@ -1955,7 +1953,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
>  	path = btrfs_alloc_path();
>  	if (!path) {
>  		extent_clear_unlock_delalloc(inode, start, end, locked_page,
> -					     EXTENT_LOCKED | EXTENT_DELALLOC |
> +					     EXTENT_DELALLOC |
>  					     EXTENT_DO_ACCOUNTING |
>  					     EXTENT_DEFRAG, PAGE_UNLOCK |
>  					     PAGE_START_WRITEBACK |
> @@ -2169,7 +2167,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
>  						      nocow_args.num_bytes);
>  
>  		extent_clear_unlock_delalloc(inode, cur_offset, nocow_end,
> -					     locked_page, EXTENT_LOCKED |
> +					     locked_page,
>  					     EXTENT_DELALLOC |
>  					     EXTENT_CLEAR_DATA_RESV,
>  					     PAGE_UNLOCK | PAGE_SET_ORDERED);
> @@ -2205,7 +2203,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
>  
>  	if (ret && cur_offset < end)
>  		extent_clear_unlock_delalloc(inode, cur_offset, end,
> -					     locked_page, EXTENT_LOCKED |
> +					     locked_page,
>  					     EXTENT_DELALLOC | EXTENT_DEFRAG |
>  					     EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
>  					     PAGE_START_WRITEBACK |
> @@ -8223,7 +8221,28 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
>  static int btrfs_writepages(struct address_space *mapping,
>  			    struct writeback_control *wbc)
>  {
> -	return extent_writepages(mapping, wbc);
> +	u64 start, end;
> +	struct inode *inode = mapping->host;
> +	struct extent_state *cached = NULL;
> +	int ret;
> +	u64 isize = round_up(i_size_read(inode), PAGE_SIZE) - 1;
> +
> +	if (wbc->range_cyclic) {
> +		start = mapping->writeback_index << PAGE_SHIFT;
> +		end = isize;

I had to go look, but ->range_cyclic can be used by the background flushing
thread, which of course will be single threaded.  However we can also trigger
this with memcg flushing during balance_dirty_pages(), which means that
writeback_index can change.  Additionally isize can change arbitrarily between
now and extent_writepages.

I think you're going to have to change extent_writepages to take the start and
end index in order to make sure we're not writing things we haven't locked.

Alternatively you could keep track of range_cyclic here, update the wbc with the
range_start and range_end that you calculate here, and reset it on the way out.

Personally I think it would be cleaner to pass in the start and end into
extent_writepages, and move the ->writeback_index update into this function, as
well as the logic for start and end.  Thanks,

Josef
diff mbox series

Patch

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 92068e4ff9c3..42bae149f923 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -464,15 +464,10 @@  noinline_for_stack bool find_lock_delalloc_range(struct inode *inode,
 		}
 	}
 
-	/* step three, lock the state bits for the whole range */
-	lock_extent(tree, delalloc_start, delalloc_end, &cached_state);
-
 	/* then test to make sure it is all still delalloc */
 	ret = test_range_bit(tree, delalloc_start, delalloc_end,
 			     EXTENT_DELALLOC, 1, cached_state);
 	if (!ret) {
-		unlock_extent(tree, delalloc_start, delalloc_end,
-			      &cached_state);
 		__unlock_for_delalloc(inode, locked_page,
 			      delalloc_start, delalloc_end);
 		cond_resched();
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 4bfa51871ddc..92726831dd5d 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -992,7 +992,6 @@  static int submit_one_async_extent(struct btrfs_inode *inode,
 				   struct async_extent *async_extent,
 				   u64 *alloc_hint)
 {
-	struct extent_io_tree *io_tree = &inode->io_tree;
 	struct btrfs_root *root = inode->root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_key ins;
@@ -1013,7 +1012,6 @@  static int submit_one_async_extent(struct btrfs_inode *inode,
 		if (!(start >= locked_page_end || end <= locked_page_start))
 			locked_page = async_chunk->locked_page;
 	}
-	lock_extent(io_tree, start, end, NULL);
 
 	/* We have fall back to uncompressed write */
 	if (!async_extent->pages)
@@ -1067,7 +1065,7 @@  static int submit_one_async_extent(struct btrfs_inode *inode,
 
 	/* Clear dirty, set writeback and unlock the pages. */
 	extent_clear_unlock_delalloc(inode, start, end,
-			NULL, EXTENT_LOCKED | EXTENT_DELALLOC,
+			NULL, EXTENT_DELALLOC,
 			PAGE_UNLOCK | PAGE_START_WRITEBACK);
 	if (btrfs_submit_compressed_write(inode, start,	/* file_offset */
 			    async_extent->ram_size,	/* num_bytes */
@@ -1095,7 +1093,7 @@  static int submit_one_async_extent(struct btrfs_inode *inode,
 	btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1);
 out_free:
 	extent_clear_unlock_delalloc(inode, start, end,
-				     NULL, EXTENT_LOCKED | EXTENT_DELALLOC |
+				     NULL, EXTENT_DELALLOC |
 				     EXTENT_DELALLOC_NEW |
 				     EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING,
 				     PAGE_UNLOCK | PAGE_START_WRITEBACK |
@@ -1263,7 +1261,7 @@  static noinline int cow_file_range(struct btrfs_inode *inode,
 			 */
 			extent_clear_unlock_delalloc(inode, start, end,
 				     locked_page,
-				     EXTENT_LOCKED | EXTENT_DELALLOC |
+				     EXTENT_DELALLOC |
 				     EXTENT_DELALLOC_NEW | EXTENT_DEFRAG |
 				     EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
 				     PAGE_START_WRITEBACK | PAGE_END_WRITEBACK);
@@ -1374,7 +1372,7 @@  static noinline int cow_file_range(struct btrfs_inode *inode,
 
 		extent_clear_unlock_delalloc(inode, start, start + ram_size - 1,
 					     locked_page,
-					     EXTENT_LOCKED | EXTENT_DELALLOC,
+					     EXTENT_DELALLOC,
 					     page_ops);
 		if (num_bytes < cur_alloc_size)
 			num_bytes = 0;
@@ -1425,7 +1423,7 @@  static noinline int cow_file_range(struct btrfs_inode *inode,
 	 * We process each region below.
 	 */
 
-	clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
+	clear_bits = EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
 		EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV;
 	page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK;
 
@@ -1575,7 +1573,7 @@  static int cow_file_range_async(struct btrfs_inode *inode,
 	memalloc_nofs_restore(nofs_flag);
 
 	if (!ctx) {
-		unsigned clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC |
+		unsigned clear_bits = EXTENT_DELALLOC |
 			EXTENT_DELALLOC_NEW | EXTENT_DEFRAG |
 			EXTENT_DO_ACCOUNTING;
 		unsigned long page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK |
@@ -1955,7 +1953,7 @@  static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 	path = btrfs_alloc_path();
 	if (!path) {
 		extent_clear_unlock_delalloc(inode, start, end, locked_page,
-					     EXTENT_LOCKED | EXTENT_DELALLOC |
+					     EXTENT_DELALLOC |
 					     EXTENT_DO_ACCOUNTING |
 					     EXTENT_DEFRAG, PAGE_UNLOCK |
 					     PAGE_START_WRITEBACK |
@@ -2169,7 +2167,7 @@  static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 						      nocow_args.num_bytes);
 
 		extent_clear_unlock_delalloc(inode, cur_offset, nocow_end,
-					     locked_page, EXTENT_LOCKED |
+					     locked_page,
 					     EXTENT_DELALLOC |
 					     EXTENT_CLEAR_DATA_RESV,
 					     PAGE_UNLOCK | PAGE_SET_ORDERED);
@@ -2205,7 +2203,7 @@  static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 
 	if (ret && cur_offset < end)
 		extent_clear_unlock_delalloc(inode, cur_offset, end,
-					     locked_page, EXTENT_LOCKED |
+					     locked_page,
 					     EXTENT_DELALLOC | EXTENT_DEFRAG |
 					     EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
 					     PAGE_START_WRITEBACK |
@@ -8223,7 +8221,28 @@  static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 static int btrfs_writepages(struct address_space *mapping,
 			    struct writeback_control *wbc)
 {
-	return extent_writepages(mapping, wbc);
+	u64 start, end;
+	struct inode *inode = mapping->host;
+	struct extent_state *cached = NULL;
+	int ret;
+	u64 isize = round_up(i_size_read(inode), PAGE_SIZE) - 1;
+
+	if (wbc->range_cyclic) {
+		start = mapping->writeback_index << PAGE_SHIFT;
+		end = isize;
+	} else {
+		start = round_down(wbc->range_start, PAGE_SIZE);
+		end = round_up(wbc->range_end, PAGE_SIZE) - 1;
+		end = min(isize, end);
+	}
+
+	if (start >= end)
+		return 0;
+
+	lock_extent(&BTRFS_I(inode)->io_tree, start, end, &cached);
+	ret = extent_writepages(mapping, wbc);
+	unlock_extent(&BTRFS_I(inode)->io_tree, start, end, &cached);
+	return ret;
 }
 
 static void btrfs_readahead(struct readahead_control *rac)