diff mbox series

btrfs: Calculate num_pages, reserve_bytes once in btrfs_buffered_write()

Message ID 20200925203638.28890-1-rgoldwyn@suse.de
State New
Headers show
Series btrfs: Calculate num_pages, reserve_bytes once in btrfs_buffered_write() | expand

Commit Message

Goldwyn Rodrigues Sept. 25, 2020, 8:36 p.m. UTC
From: Goldwyn Rodrigues <rgoldwyn@suse.com>

Cleanup.

write_bytes can change in btrfs_check_nocow_lock(). Calculate variables
such as num_pages and reserve_bytes once we are sure of the value of
write_bytes so there is no need to re-calculate.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
---
 fs/btrfs/file.c | 34 ++++++++++++----------------------
 1 file changed, 12 insertions(+), 22 deletions(-)

Comments

Josef Bacik Oct. 5, 2020, 1:10 p.m. UTC | #1
On 9/25/20 4:36 PM, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Cleanup.
> 
> write_bytes can change in btrfs_check_nocow_lock(). Calculate variables
> such as num_pages and reserve_bytes once we are sure of the value of
> write_bytes so there is no need to re-calculate.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Reviewed-by: Josef Bacik <josef@toxicpanda.com>

Thanks,

Josef
David Sterba Oct. 14, 2020, 1:08 p.m. UTC | #2
On Fri, Sep 25, 2020 at 03:36:38PM -0500, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Cleanup.
> 
> write_bytes can change in btrfs_check_nocow_lock(). Calculate variables
> such as num_pages and reserve_bytes once we are sure of the value of
> write_bytes so there is no need to re-calculate.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>

Added to misc-next, thanks.
diff mbox series

Patch

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 038e0afaf3d0..7695988fdba3 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1648,8 +1648,7 @@  static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		size_t write_bytes = min(iov_iter_count(i),
 					 nrptrs * (size_t)PAGE_SIZE -
 					 offset);
-		size_t num_pages = DIV_ROUND_UP(write_bytes + offset,
-						PAGE_SIZE);
+		size_t num_pages;
 		size_t reserve_bytes;
 		size_t dirty_pages;
 		size_t copied;
@@ -1657,8 +1656,6 @@  static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		size_t num_sectors;
 		int extents_locked;
 
-		WARN_ON(num_pages > nrptrs);
-
 		/*
 		 * Fault pages before locking them in prepare_pages
 		 * to avoid recursive lock
@@ -1670,35 +1667,28 @@  static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 
 		only_release_metadata = false;
 		sector_offset = pos & (fs_info->sectorsize - 1);
-		reserve_bytes = round_up(write_bytes + sector_offset,
-				fs_info->sectorsize);
 
 		extent_changeset_release(data_reserved);
 		ret = btrfs_check_data_free_space(BTRFS_I(inode),
 						  &data_reserved, pos,
 						  write_bytes);
 		if (ret < 0) {
+			/*
+			 * If we don't have to cow at the offset, reserve
+			 * metadata only. write_bytes may get smaller
+			 * than requested here.
+			 */
 			if (btrfs_check_nocow_lock(BTRFS_I(inode), pos,
-						   &write_bytes) > 0) {
-				/*
-				 * For nodata cow case, no need to reserve
-				 * data space.
-				 */
+						   &write_bytes) > 0)
 				only_release_metadata = true;
-				/*
-				 * our prealloc extent may be smaller than
-				 * write_bytes, so scale down.
-				 */
-				num_pages = DIV_ROUND_UP(write_bytes + offset,
-							 PAGE_SIZE);
-				reserve_bytes = round_up(write_bytes +
-							 sector_offset,
-							 fs_info->sectorsize);
-			} else {
+			else
 				break;
-			}
 		}
 
+		num_pages = DIV_ROUND_UP(write_bytes + offset, PAGE_SIZE);
+		WARN_ON(num_pages > nrptrs);
+		reserve_bytes = round_up(write_bytes + sector_offset,
+					 fs_info->sectorsize);
 		WARN_ON(reserve_bytes == 0);
 		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
 				reserve_bytes);